Home

HP LeftHand Storage with VMware vSphere: Design considerations

Contents

1. Virtual Network Adapter WMkernel mki Switch wSwitchi Port Group YMkernel For CST Motion 4 Port Group Policy g Compliant IF Address 192 168 0 135 Subnet Mask 255 200 25o8 Physical Network Adapter Mame vrninice Device Emulex Corporation OneConneck 1066 MIC bes Link Status Connected Configured Speed 1000 Mbps Full Duplex iSCSI HBA connectivity An iSCSI host bus adapter iSCSI HBA uses dedicated hardware to connect to a 1GbE or 10GbE network while mitigating the overhead of iSCSI and TCP processing of Ethernet interrupts and thereby improves the performance of servers iSCSI HBAs can include support for booting from an iSCSI target Storage connectivity via iSCSI HBAs therefore enables both offloading of the iSCSI processing from the server and booting from HP LeftHand Storage iSCSI HBAs do not require licensing or special networking within vSphere 5 hosts as they provide a dedicated storage connection for iSCSI targets only and are managed within vSphere 5 similarly to Fibre Channel HBAs As best practice use two iSCSI HBA initiators a dual port adapter or two single port adapters each configured with a path to all iSCSI targets for failover Configuring multiple iSCSI HBA initiators to connect to the same target requires each iSCSI HBA initiator s IQN to be configured as server connection on the HP LeftHand Management Group This is configured in the HP LeftHand Centralized Management Console as two server co
2. Four network adapters two dedicated to iSCSI Networking Refresh Standard Switch vSwitcho Remove Properties eee natsaa Pen 4 YM Network BA vmnici 1000 Full A El 2 virtual machines EA vronicd 1000 Full A P4000 M WIN2 a P4000 SIM ah WMkemel Port LJ Management yvMotion Network a vmk 172 16 44 238 Standard Switch vowitchd Remove Hepes Peters Sie elas Se E a a1 eae T 4 YMkernel For i551 B BA vronics 1000 Full A vook2 192 168 0 138 GA vronice 1000 Full 4 Mkemel Port 4 VMkernel For iSCSI A vmki 192 168 0 137 Six network ports vSphere 5 hosts with six Gigabit network ports are ideal for delivering performance with the iSCSI software adapter The improvement over four ports is achieved by separating vMotion and FT traffic from management traffic so that they do not have to share bandwidth Each traffic type performs better in this environment To configure vSphere 5 hosts with six Gigabit network ports use three standard virtual switches each comprising two Gigabit ports as shown in figure 11 If possible one port from each of the separate Gigabit adapters are used in each team to prevent some bus or network interface card failures from affecting an entire standard virtual switch 11 e The first standard virtual switch should have AVMkernel port network with management traffic enabled with the two network adapters for load balancing default and avirtual machine port group network
3. HP LeftHand Storage Figure 4 HP Insight Control for VMware vCenter Server with monitoring for servers storage and Virtual Connect networking oro co Susan_VM01 ERE ee E casl adap ETBUR E ria a gt ae ETSER Getting Started Summary ATT a e a Performance Configuration Tasks amp Events Alarms Permissions A a A HP Insig HP Management Status Summary Server Infrastructure Refresh Properties gt Gian Health Reported by iLO tg qes E Server Management System Heath E Configuration a a EES TIEA EREE IE EER A Power Thermal Blade Status O0 Networking Firmware Roweres en Power Allocated 248 Logs Storage Information Health Reported by HP Systems Insight Manager Storage Overview Health Status OOO Related VMs Management Processor Status O00 Related Datastores Related Hosts Health Reported by HP Virtual Connect Manager Storage Tools Health Status O00 J About Health Reported by HP Storage Management Hewlett Packard Development HP P9000 Storage Array S3031 O0 Company L P HP XP Disk Array 80047 O O0 Plug in Version information HP XP Disk Array 80207 O00 Server Module 6 3 0 0 0 Storage Information 6 3 0 HP P6000 EVA Calamity_F HP MSA2000 G2 David MSA G2 O0 2010 All rights reserved HP P6000 EVA Hurricane_A O O HP P4000 cd mg1 O0 iLO Remote Management 10 10 9 148 Web Administration Access the iLO web user interface
4. SmartClone volume contains the changes from the temporary space and point to the source snapshot In order to preserve data on the volume which was written prior to conversion a SmartClone the volume needs to be disconnected from the vSphere 5 host s Alternatively selecting the Delete Temp Space option from a context menu of the snapshot in the CMC will forcibly remove this data from HP LeftHand Storage Once temporary space is deleted it may not be recovered It is a best practice to unmount the snapshot from any vSphere 5 hosts prior to deleting temporary space If you wish to remount the snapshot you can issue the Rescan All command in the vSphere Client prior to mounting the snapshot again Snapshots offer excellent ways for promoting data recovery 26 Figure 25 Converting Temp Space into a SmartClone volume using the CMC E HP Soret E00 Gene Managenent Conse oe Filey Find Tasks WHelpy Check for Upgrades z Getting Started Eir neadan mii Details Remote Snapshots Assigned Servers iSCSI Sessions Map View a pe 5 ry Local Snapshot Events f Servers 1 Name E Volume_SS_1 A Administrati wag Administration Description Sites E FortCollins Cluster FortCollins Performance Monitor P Status N Temp space Normal gt Storage Systems 1 on ormal Temp sp H amp Volumes 1 and Snapshots 1 Server ESxi2 E lt Gio Volume 1 i LLB Volume_ Primary l New SmartClone Vo
5. streamlined patching and configuration model e Storage Distributed Resource Scheduler Storage DRS aggregates storage into pools simplifies scale management ensures best VM load balancing and helps avoid storage resource bottlenecks e Profile driven storage matches appropriate storage to given VMs thus ensuring correct placement e Virtual Machine File System version 5 VMFS 5 enhances scalability and increased performance supporting larger 64 TB datastores a unified 1 MB block size and small file space utilization with 8 KB sub block sizes e Improved vSphere Storage APIs for Array Integration VAAI and new vSphere Storage APIs for Storage Awareness VASA enhance performance and bring array information into DRS and profile driven features HP LeftHand Storage increases and builds upon features found in HP LeftHand operating system HP LeftHand OS formerly known as SAN iQ version 9 0 with version 9 5 e VAAI introduced in vSphere 4 1 improves storage efficiency and performance Note that newer VAAI feature extensions introduced in vSphere 5 0 are not supported in HP LeftHand OS 9 5 e HP LeftHand OS 9 5 expands application integration with Application managed Snapshots for VMware Application consistent point in time copies allow speedy and reliable Virtual Machine VM recovery e HP LeftHand OS 9 5 features Zero to VSA assisted and automated installation for easy deployment of HP LeftHand Storage Virtual SAN Appliance V
6. Enabled ais Site ELE 2 IP Address Model Manager Special Manag RAID Status RAID Corni Servers 1 Description Controlling Se iSCSI Mode Initiator Node CHAP Name Load Bala Lm P4000 ESx2 172 16 44 232 No CHAP required ign 1998 01 com Enabled Unassigned Creating the first iSCSI volume Before a vSphere 5 host can create a new VMFS datastore on an HP LeftHand volume that volume must be created on the HP LeftHand cluster and must have been configured to allow the host to access the volume All hosts within a vSphere 5 cluster must also have access to that same volume in order to support vMotion High Availability and Fault Tolerance Use the CMC as shown in figure 16 or use the command line interface CLI to create a new volume Then using the CMC create a new server connection with the vSphere 5 host s IQN as configured for the iSCSI software adapter optionally use CHAP secrets as required Please refer to the HP LeftHand User Guide for more information Figure 16 Creating anew volume on HP LeftHand Storage iz New Volume Basic Advanced Volume Name ES Volume Cluster Available Space 3 74755 TB Reported Size oT Discovery of the iSCSI volumes To discover volumes the iSCSI software adapter of each vSphere 5 host must have added the Virtual IP VIP address of the HP LeftHand cluster which contains the volume in its Dynamic Discovery Send Targets list see figure 13 New targets can then be discove
7. Storage to make full copies of data within the HP LeftHand storage clusters Previously data movement would traverse from the storage arrays occupying bandwidth to the vSphere host utilize CPU resources processing the data movement and again traverse back to the storage array 30 e Block zeroing Enables HP LeftHand Storage to zero out large numbers of blocks with a single command Previously repetitive zeros would traverse from the vSphere host to the storage arrays occupying bandwidth and utilizing CPU resources processing the data movement This feature only speeds up the process of provisioning of VMs created with the eager zeroed thick disks option which is not the default but required for Fault Tolerance e Hardware assisted locking Atomic Test and Set Provides a new efficient mechanism to protect metadata for VMFS in cluster deployments improves scalability of large vSphere host farms sharing a VMFS datastore and is more efficient than the previous SCSI reserve and release mechanism HP LeftHand volumes are shared across multiple clustered vSphere hosts Improved efficiency in sharing volumes across the cluster directly impacts the number of VMs a volume can hold increases the size of the volume to support the increased VM count and increases the number of hosts within a cluster sharing access to the same LUN Figure 29 VAAI performance improvements with HP LeftHand Storage Time to complete 1400 1200 1000 800 E without off
8. adapters For this port group In a failover situation standby adapters activate in the order specified below Name Speed Networks Name Speed Networks Move Up Active Adapters Active Adapters 192 168 0 229 192 168 0229 ER vmnic2 1000 Full 192 168 0 229 192 1668 0 229 g vmnic3 1000 Full Standby Adapters Standby Adapters Unused Adapters Unused Adapters GB vmnic3 1000 Full 192 168 0 229 192 168 0 229 GB vmnic2 1000 Full 192 168 0 229 192 166 0 229 Adapter Details Adapter Details Name Name Location Location Driver Driver Cancel Help Cancel Help As a best performance practice the VMkernel ports for iSCSI should be separate from the host management and virtual networks used by virtual machines However high availability is usually preferred over splitting networks on hosts with a small number of network adapters If enough networks are available to provide network redundancy it is ideal to separate vSphere vMotion and Fault Tolerance FT networks on unique virtual switches These networks are placed on separate vSphere standard switches within each host or distributed switches which handle networking traffic for all associated hosts on the datacenter Port groups defined with unique network labels also help to organize and separate traffic In a vSphere distributed switch with network I O control enabled distributed switch tr
9. are the best way to deploy small quantities of cloned golden image VMs or virtual desktops Manageability and deployment of VDI environments are best to leverage VMware View implementations with Linked Clones Please refer to the HP VDI Reference Architectures 28 Figure 27 Map view of HP LeftHand SmartClone volumes in the CMC aa 15 OS 16 AS Ae Custom lmagez Custom Apps Oo 14 i Am O54 O58 1 Snapshot D AS p g Gold_Image SMEG aSr o T m Custom Image O56 O8 12 T g5 J527 OS 11 Ay im Custom Apps OS 10 T 013 SmartClone volumes can be used seamlessly with any other HP LeftHand feature including snapshots or Remote Copy SmartClone volumes are also very useful for performing tests on virtual machines by quickly reproducing them without actually using space on the HP LeftHand cluster to copy them Unlike temporarily mounted snapshots SmartClones are meant to stay on the HP LeftHand cluster for a longer time and any delta data is persisted in the SmartClone volume Please see the section on Resignaturing snapshots Note that every write including a delete is a delta block persisted in the SmartClone delta space If long term space efficiency is required stick to the same best practices as for thinly provisioned volumes such as minimizing writes to VMFS datastores on a SmartClone volume including avoiding defragmentation within the guest VM Successful approaches have also included separation of User and Application data inclu
10. for storing application data typically for separation from a virtual machine s boot volume Specifically as a best practice database and log volumes for many applications require separate volumes These should be presented as either RDMs through your vSphere 5 hosts or connected as iSCSI disks directly through the iSCSI initiator software of the virtual machine s guest operating system direct iSCSI Using RDMs or direct iSCSI allows these application volumes to be transported seamlessly between physical and virtual servers as they are formatted in the native file system NTFS EXT3 etc of the guest operating system To use the guest operating system initiator successfully follow these guidelines e Make sure the guest operating system initiator is supported by HP LeftHand Storage Please check the compatibility listing for HP LeftHand Storage on HP SPOCK e For optimal performance and failover the virtual machine network used for iSCSI by the guest s iSCSI initiator should have at least dual Gigabit links and be separate from other virtual networks including VMkernel instances for vMotion Fault Tolerance logging host management and virtual machine public networks e Make sure that the guest operating system is using the para virtualized VMXNET 3 network adapter driver which is typically installed through VMware Tools on supported guests e When using direct iSCSI make sure that the virtual machine will not be used in combination
11. immediately figure 22 Go to Storage and select the VMFS datastore which should be expanded Click on Properties and choose increase Select the same HP LeftHand Storage volume which reports the changed size The VMFS datastore will be increase by using the free space on the volume oe oS Se E Figure 22 vSphere Client now reflects the change made to the iSCSI volume CFT144 vSphere Client File Edit View Inventory Administration Plug ins Help gt a Home gt 2S Inventory p if Hosts and Clusters aa earch Inventor a ae x B 192 168 1 25 VMware ESXi 5 0 0 441354 CFT144 By Datacenter Getting Start Summary i Resource ation e em Configuration Nay amp Everts Permissions Maps Storage Views Har e Status 5 g 13 13 13 11 Started Virtual Machines Alloc Perf onfi asks Alarms dware St a vMo01 itardiovare Storage Adapters Refresh Rescan All S yMo02 L J g vMoos Processors E vos iiil oo g vmoos panii ign 1998 01 com vmware bay 1 7 1f924ee a Networking 7 wor a VMOO B vmbbaz Block SCSI vM008 x 1SP2432 based 4Gb Fibre Channel to PCI Express HBA gt wae svete ne seme vmhbao Fibre Channel 50 01 43 80 03 c0 64 29 0 01 43 80 03 c0 64 28 a ce Advanced Settings AA umhhat Fikra Channel er e e te T C T OT xi Power Management Details amp vmoi2 L vM013 Softy Sw _ _ ew fl Model ISCSI Software Adapter a E 13 13 13 12 Licensed Foatures iSCSI Name ign 1998 0
12. increase available bandwidth or add more ports for any other network services such as virtual machine networking as required When adding more ports for iSCSI be sure to create additional VMkernel ports for iSCSI map them to one physical adapter on the vSwitch and add them to the iSCSI software initiator 10GbE options Both vSphere 5 hosts and HP LeftHand Storage support 10GbE interfaces and benefit from increased bandwidth 10GbE network ports offer additional bandwidth over 1GbE interfaces With HP LeftHand supporting 10GbE interface upgrades throughput has been seen to increase 2 3x depending upon workload patterns HP LeftHand cluster size and networking infrastructure to the hosts Note that in 10GbE networks full mesh backplanes and latency options vary and impact performance Flexible network infrastructure tailors variable bandwidth options to need Generally storage performance for small block random workloads is limited by the number of disk spindles dictating input output operations per second IOPS characteristics capabilities of the physical storage In scalable storage architectures like HP LeftHand Storage workload is distributed across the storage cluster Note that in all previous examples 1GbE could have easily been replaced by faster network bandwidth options doing so will impact best practice choices It is best practice to balance networking technologies end to end for overall solution performance characteristi
13. naa 6000eb310a4F20ca0000000000004e14 Manage Paths Policy Path Selection Round Robin Cy Mware Change Storage Array Type Vil SATP_ DEFAULT 44 Paths Runtime Mame Target LON Status ymhbasa c T7 L0 ign 2003 10 com lefthandnetworks p4800 vdi mg 19988 O J Active Tio wroihbas2 CiUiTALO ign 2003 10 com lefthandnetworks p4s00 vdi mg 19985 0 Active Io Freferred Refresh Mame ign 1998 01 com vmware wdi esx2 vdi local O0023d000001 ign 2003 10 com lefthandnetworks p4800 wdi mg 19958 Runtime Mame vmhbasz c0 T7 L0 SCSI Adapter ign 1998 01 com vinware yvdi esx2 vdi local iSCSI Alias Target iqn 2003 10 com lefthandnetworks p4800 vdi mg 19985 permanentdesktops 1 7 152529 192 168 07 3260 Alternative In the host s Configuration Storage highlight the VMFS datastore for which you want to manage the paths right click select Properties and select the Manage Paths option e From the host s Configuration Storage highlight the VMFS datastore Note in Datastore Details the number of Total Paths should accurately reflect the number of valid paths to the storage volume For each network adapter used by the iSCSI software adapter an individual iSCSI session is established to HP LeftHand Storage The example referred to in figure 14 gives each HP LeftHand volume two iSCSI paths using two separate physical network adapters The round robin policy ensure
14. that the cluster virtual IF be configured Authentication CHAP not reguireg CHAP required When creating the Application managed Snapshot of a volume that contains a VMFS datastore the CMC communicates with vCenter Server during the snapshot process vCenter Server quiesces VMs and takes VMware Snapshots VMware Tools 24 within each guest VM may quiesce applications inside the VMs during this process Once all VMware Snapshots are taken HP LeftHand Storage performs a snapshot of the volume which now contains VMware Snapshots on the HP LeftHand storage cluster This can be seen when mounting the snapshot to a vSphere 5 host After the snapshot of the VMFS datastore volume has been created on HP LeftHand Storage the vSphere VM snapshots are removed Resignaturing snapshots In order for vSphere 5 hosts to utilize SAN based snapshots or SmartClones a snapshot or a SmartClone of a volume is taken and presented to the vSphere hosts The process of presenting a snapshot or SmartClone volume is the same as assigning volumes to a server connection in the CMC To connect to the newly assigned snapshots or SmartClones rescan the iSCSI software adapters on the vSphere hosts vSphere 5 supports three options when mounting a HP LeftHand snapshot or SmartClone volume that has VMFS data Keep the existing signature assign a new signature or format the disk figure 24 e Keep the existing signature allows you to mount the VMFS datastore
15. unique iSCSI session Since HP LeftHand Storage scales by leveraging multiple storage nodes compute disk and network resources iSCSI multi pathing s advantage is seen in multi threading data approach up to the maximum throughput sizing capability of the HP LeftHand cluster Configuring iSCSI multi pathing requires at least two network adapters on the iSCSI network The following steps are performed on each vSphere 5 host individually e From the host s Configuration Networking select Properties of the standard virtual switch Select the Network Adapters tab and add the appropriate adapters on the iSCSI network e Select the Ports tab and Edit or Add VMkernel Ports Configure as many VMkernel ports as network adapters used for iSCSI On the VMkernel Properties NIC Teaming tab select one physical adapter vmnic to be listed as Active Adapter all other physical adapters should be listed as Unused Adapter not Standby Adapter Repeat for remaining VMkernel ports and select another physical adapter to be the active adapter such that each VMKernel port has only one exclusive physical adapter assigned to it See the following example for two NICs and two VMkernel ports figure 8 Port group VMkernel port name Active adapter Unused adapter iSCSI A vmnic2 vmnic3 iSCSI B vmnic3 vmnic2 e From the host s Configuration Storage Adapter highlight the iSCSI Software Initiator In the Network Configuration tab add all VMker
16. without changing the signature e Assign a new signature allows you to retain the existing data and mount the VMFS datastore present on the disk The signature will change and uniquely identify the volume without impacting data within the VMFS file system e Format the disk will destroy all data on the volume Keeping the existing signature during a mount operation may generate an error when there is conflict with an already mounted VMFS datastore on the host or cluster that has the same UUID VMFS signature If the original volume that contains the original VMFS datastore and its snapshot is mounted at the same time in a vSphere datacenter force mounting the VMFS datastore from the snapshot LUN on this same vSphere 5 host is not allowed In this case the only way to mount that VMFS datastore from the snapshot LUN is to assign a new signature and choose a new label for that VMFS datastore By assigning a new signature references to the existing signature from virtual machine configuration files will need to be updated Note in the Ready to Complete summary during the VMFS datastore mount process the Original UUID signature is provided along with Assign a new UUID The resulting event task is resignature unresolved VMFS datastore With a successful mount the VMFS datastore mounts with a newly generated Identification snap XXXXXXXX lt Original label gt On volume properties select Rename to provide a more suitable volume identification n
17. 1 com vmware bay1 71f924ee G RedHat Lin Time Configuration iSCSI Alias D DNS and Routing Connected Targets 4 Devices 2 Paths 4 D yMo1 Authentication Services RET pane amp y M018 Power Management gt vmo Virtual Machine Startup Shutdown Name Id Runtime Name LUN Type Transport Capacity Owner Hardware Accelere M020 Virtual Machine Swapfile Location LEFTHAND iSCSI Disk naa 6000eb3c1fcfbbcb nm wmbhba32 C0 TO LO 0 disk iSCSI 1 20 78 NMP Supported Security Profile LEFTHAND iSCSI Disk naa 6000eb3c1fcfbbcb n vmbba32 CO T1 L0 0 disk iSCSI 500 00G NMP Supported System Resource Allocation 4 gt a 21 Advanced Settinas T 2 xj Recent Tasks Name Target or Status contains Cex X Target Status Intiatedby Requested Start Ti Start Time F Refresh host storage system 13 13 13 11 Completed Administrator 3 14 2011 12 52 57 3 14 2011 12 53 01 3 14 2011 F Rescan vMFS 13 13 13 11 Completed Administrator 3 14 2011 12 52 52 3 14 2011 12 52 55 3 14 2011 w Snapshots Remote Copy and HP SmartClone volumes HP LeftHand Application managed Snapshots in vSphere 5 HP LeftHand OS version 9 5 expands application integration functionality to include volumes with VMware Virtual Machines HP LeftHand Application aware Snapshot Manager enables HP LeftHand snapshots for VMFS datastores thereby creating application consistent point in time states for reliable recovery of VM
18. Add Vendor Provider x Vendor Provider Vendor Provider Details Login I Use Vendor Provider Certificate Certficatelocation U ov zs ES ow Name Target Status Details Initiated by vCenter Server Requested Start Ti Start Te For additional information best practices and software please go to the HP Insight Control site http h18000 www1 hp com products servers management integration html VMware High Availability and Distributed Resource Scheduler The advanced vSphere 5 features such as vMotion Clusters High Availability HA Fault Tolerance FT and Distributed Resource Scheduler DRS require multiple vSphere 5 hosts to have simultaneous access to the same set of volumes on shared storage To enable this on a HP LeftHand volume assign multiple servers one for each vSphere 5 server iSCSI software adapter to the same volume To use multiple servers simply create one server connection for each vSphere 5 host that connects to HP LeftHand Storage and create a server cluster using the steps outlined in the section Creating the first iSCSI volume When multiple server connections are assigned to the same volume in the CMC a warning indicates that this configuration is intended for clustered servers or clustered file systems such as vSphere 5 figure 6 When assigning a server cluster rather than individual server connections this warning will not appear A server connection with iSCSI qualified name IQ
19. CSI network that IP addresses can be pinged Please refer to Discovery of the first iSCS volume Is virtual IP load balancing supported in vSphere 5 In vSphere 5 iSCSI software adapter and iSCSI HBAs are supported by HP LeftHand Virtual IP Load Balancing As a best practice load balancing should be enabled in the CMC on the server connections or server clusters of all vSphere 5 hosts This is enabled by default What initiator should use for a virtual machine s boot volume There are many ways to connect and present an iSCSI volume to a virtual machine on a vSphere 5 host These include using the iSCSI software adapter iSCSI HBAs or the iSCSI initiator provided by guest operating systems Please check the compatibility listing for HP LeftHand Storage on HP SPOCK for information on iSCSI initiator support For VMFS datastores containing a virtual machine s definitions and virtual disk files you need to use the iSCSI software adapter or an iSCSI HBA on the vSphere 5 host You may use either adapter Both give you full vSphere 5 functionality and both are supported equally Using iSCSI HBAs has the advantage of offloading iSCSI processing from the vSphere 5 host Note that although vSphere 5 supports iSCSI Boot Firmware Table iBFT it is currently not supported or listed in HP SPOCK or the HP LeftHand Compatibility Matrix What initiator should I use for additional raw device mappings or virtual disks Additional volumes are used
20. Integrated Remote Console Access the system KYM Requires Active x Remote Console Access the system KYM Requires JRE gt Remote Serial Console Access a T320 serial console connected to Virtual Serial Port xl The VASA interface provides information about the Storage DRS management capability for VMFS datastores and volumes A VASA provider can indicate that the storage device supports Storage DRS migration and whether migration between VMFS datastores is recommended Figure 5 Registering the VASA provider for HP LeftHand Storage ernan r Ble gc OT A A A a e a T E ee ee i A N A T ee aa a e AAT HP Insight Software HP Management Refresh in progress Overview A refresh is currently in progress Data will be refreshed in the background Server Management Elapsed time 0 minutes 27 seconds Estimated time remaining 0 minutes 10 seconds Check Refresh Progress Configuration Power Thermal About ge Networking Storage Setup for HP Insight Control Storage Module for vCenter Last updated 23 hours 12 minutes ago Refresh B Firmware Storage Provider js Logs https na casidev casl adapps hp com 3711 vasa_provider_ws vasaService Storage Information Storage Overview Configured Arrays and Management Servers Launch Storage Administrator Portal Related VMs File Edt View Inventory Administration Plug ins Help Ce ee Se Remove Refresh Al Sync Vendor Providers Name URL Last Refresh Time Last Sync Time
21. MFS 3 to VMFS 5 is non disruptive and immediate however volumes will still retain their original block size Therefore modifying block size to fully recognize new VMFS 5 benefits will require a reformat of the volume This is particularly problematic if the data needs to be retained or VMs depending upon the volume need to stay online It is therefore recommended to create new volumes on HP LeftHand Storage as a proposed upgrade replacement Storage vMotion would then be leveraged for moving current VMs from VMFS 3 to VMFS 5 datastores Additional increased efficiency is achieved with sub block addressing reducing the storage overhead associated with smaller files Small files need not occupy an entire 1 MB unified block as sub blocks of 8 KB and 1 KB are now possible Raw Device Mappings RDMs larger than 2 TB are now also supported Note that VMFS 5 has finer grained SCSI locking controls supporting hardware assisted locking in VAAI In order to take full advantage of this feature found in HP LeftHand OS version 9 0 and above ensure that the VMFS datastores are using VMFS 5 Hardware assisted locking improves scalability and performance The lock manager for VMFS 5 scales linearly with the number of vSphere hosts in the cluster resource pool which is particularly useful in conjunction with the scalability characteristics of HP LeftHand Storage This impacts the number of virtual machines per VMFS datastore the overall storage scalability and helps to f
22. N authentication maps one vSphere 5 iSCSI software adapter to one or more volumes In addition Challenge Authentication Protocol CHAP authentication can also be used for increased security either one way target secret or two way target and initiator secret Each vSphere 5 host must be configured to use the correct CHAP credentials New volumes on each vSphere 5 host must be discovered Dynamic or Static Discovery as described in the section Discovery of the first iSCS volume Figure 6 Warning when assigning a volume to multiple server connections Centralized Management Console You are inthe process of assigning more than one server to a volume or snapshot Best practices recommend using a server cluster Associating more than one server to a volume or snapshot is intended for clustered servers or clustered file systems To continue the process and associate more than one server to this volume or snapshot click Continue To return to the previous window or to create or edit a server cluster click Cancel lS Exe Initial iSCSI setup of vSphere 5 Networking for the iSCSI software adapter Before storage connectivity via the iSCSI software adapter is enabled in vSphere 5 a specific network configuration is recommended For best connectivity use at least two 1GbE or 10GbE network adapters for performance and failover figure 7 on the iSCSI network which connects hosts running vSphere 5 0 to HP LeftHand Storage Figure 7 vS
23. SA e HP LeftHand OS 9 5 supports new Storage Replication Adapter SRA for VMware vCenter Site Recovery Manager 5 0 Site Recovery Manager 5 SRM offering integrated storage failover and failback e Network RAID 5 and 6 dynamic stripe size improves efficiency and is applied to the entire volume stack Scheduled snapshots for space efficiency are no longer required e Successfully addressing and deploying these new features across the solution is imperative if you wish to maximize the return on investment ROI for HP LeftHand Storage and VMware vSphere while continuing to meet the dynamic needs of the business To help you successfully deploy these new features and maximize the performance and scalability of your vSphere environment this paper presents best practices for configuring and tuning vSphere 5 storage enhancements for HP LeftHand Storage vSphere 5 offers many new capabilities over vSphere 4 1 improving and extending storage specific benefits with the integration of HP LeftHand Storage These new features and enhancements provide increased performance optimization easier provisioning monitoring and troubleshooting thereby impacting utilization and operational efficiency VMFS 5 VMFS 5 introduces core architectural changes enabling greater scalability and performance Major features introduced include 64 TB device support unified block size and improved sub blocking The unified block size is now 1 MB Note that upgrading V
24. Tasks a Expanding a VFMS datastore 1 Connect to vCenter and select one vSphere 5 host to which the iSCSI LUN is presented 2 Click on the Configuration tab and select Storage Adapters 3 Highlight the iSCSI software adapter and locate the volume in the Details section The LUN is still showing 1 TB figure 21 22 Figure 21 vSphere Client shows iSCSI volume still as 1 TB prior to Rescan All Resource Allocation Performance Gem Tasks amp Events Alarms Permissions Maps Storage Views Hardware Status Storage Adapters Refresh Rescan All Device iSCSI Software Adapter G vmhbazz ign 1998 01 com vmware bay l 71f924ee Smart Array P4101 G vmbbaz Block SCSI 15P2432 based 4Gb Fibre Channel to PCI Express HBA vmbbad Fibre Channel 50 01 43 30 03 c0 64 29 50 01 43 80 03 c0 64 28 P o umhhai Fikes Channel Seeder ee OR Seder Ad a x Details ymhba3 z Properties Model iS ST Software Adapter iS ST Mame ign 1995 01 corm viware bay1 71F924ee iS ST Alias Connected Targets 4 Devices 2 Paths 4 Tiew Devices Paths Mame Td Runtime Mame LOUN Type Transport Capacity Owner Hardware Acceler LEFTHAND iSCSI Disk inaa 6000eb3cifctbbcb o oma vmhba3z c0 T0 L0 0 disk iScsSI 1 00 TE NMP Supported LEFTHAND iSCSI Disk inaa 6000ebacifecfbbeb ma mhbas2 CO T1 LO D disk iSCSI 500 00 S NMF Unknown Right click on the iSCSI software adapter and select Rescan The new size is reported
25. Technical white paper HP LeftHand Storage with VMware vSphere Design considerations and best practices Table of contents Executive summary 3 New Storage feature 3 vSphere 5 storage enhancements for HP LeftHand Storage 4 VMFS 5 4 Storage DRS 4 HP Insight Control Storage Module for vCenter and vSphere Storage APIs for Storage Awareness VASA 6 VMware High Availability and Distributed Resource Scheduler 8 Initial iSCSI setup of vSphere 5 8 Networking for the iSCSI software adapter 8 Two network ports 10 Four network ports 11 Six network ports 11 More than six network ports 13 10GbE options 13 Enabling the iSCSI software adapter 13 iSCSI HBA connectivity 14 Multi pathing for iSCSI 15 Connecting and using iSCSI volumes 17 Enabling load balancing 17 Creating the first iSCSI volume 18 Discovery of the iSCSI volumes 18 Disconnecting iSCSI volumes from vSphere 5 hosts 19 Troubleshooting volume connectivity 19 Creating a new datastore ona volume 20 Expanding an iSCSI LUN in the CMC 20 Expanding a VFMS datastore 22 Snapshots Remote Copy and HP SmartClone volumes 24 HP LeftHand Application managed Snapshots in vSphere 5 24 Resignaturing snapshots 25 HP LeftHand Snapshots of Raw Devices 27 HP LeftHand Snapshots on VMFS datastores 27 HP LeftHand Remote Copy volumes and SRM HP LeftHand SmartClones Virtual machine placement and VAAI Choosing datastores for virtual machines vSphere Storage APIs for Array Integration VAAI Best pra
26. affic is divided across predefined network resource pools VMware FT traffic iSCSI traffic VMotion traffic management traffic NFS traffic and virtual machine traffic Note the similarities with standard switches and manual segregation However network 1 0 control enforces these rules across a datacenter thereby easing policy enforcement and management The resources that you can deploy vary based upon the type and quantity of resources available Larger HP LeftHand Storage clusters can accommodate different requirements than smaller HP LeftHand clusters HP BladeSystem c7000 solutions with HP LeftHand P4800 G2 in the same enclosure have unique benefits with internal extremely low latency network characteristics for iSCSI through HP Virtual Connect technology vs this same cluster connecting to external devices A VM s network traffic requirements differ in every application These basic guidelines help you follow best practice for connecting HP LeftHand Storage However understanding the approach to bandwidth tuning involves understanding the complete solution objectives and needs HP BladeSystem with HP Virtual Connect Flex 10 or HP Virtual Connect FlexFabric present tunable network interfaces as physical adapters within vSphere thereby offering competing bandwidth allocation and enforcement opportunities than with vSphere Management and deployment options will vary In larger multiple HP BladeSystem environments HP CloudSystem Matrix offers add
27. ame 25 Figure 24 Add Storage wizard when mounting a volume clone or snapshot Add Storage 2 _ Select VMFS Mount Options Specify if you want to mount the detected VMFS volume with the existing signature use a new signature or format the disk E Disk LUN Select Disk LUN Mount Options Keep the existing signature Ready to Complete Mount the VMFS volume without changing the signature Spedfy a VMFS mount option C Assign a new signature Retain the existing data and mount the VMFS volume present on the disk C Format the disk Create a new datastore Note that HP LeftHand snapshots are persistent points in time representing data on the source volume at the time it was taken As an extension to this feature HP LeftHand allows snapshots to be mounted and any changes are written toina temporary area This data area is called a temporary space and contains delta writes from the original snapshot data By assigning a new Signature in vSphere 5 we are writing data to this temporary space From the HP LeftHand storage cluster perspective temporary space will be lost if HP LeftHand Storage is rebooted or manually removed It retains the same high availability characteristics of the source volume If data written to the temporary space along with the snapshot needs to be preserved in a new volume the CMC may be used to create a space efficient new volume under a new name and new access permissions figure 25 The resulting
28. cs For example HP LeftHand Storage with 10GbE connectivity can be limited by using a 1GbE switching infrastructure Enabling the iSCSI software adapter The vSphere 5 iSCSI software adapter is the most common way to connect a vSphere 5 host to HP LeftHand Storage It has to be enabled on each vSphere 5 host To enable the iSCSI software adapter go to Configuration Storage Adapters to add it to the host s list of storage adapters Configuration of the iSCSI software initiator is done through each host s storage adapters list Here are some guidelines when using the software iSCSI software adapter e Enable the iSCSI software adapter on each vSphere 5 host e Make sure all VMkernel instances which are connected to the iSCSI network are assigned to the iSCSI software adapter figure 12 e Copy or write down the IQN that identifies each vSphere 5 host it will be needed for the server connection on the HP LeftHand management group later to define volume access on the storage cluster 13 Figure 12 Two VMkernel ports bound to the iSCSI software initiator iSCSI Initiator v mbba327 Properties General Network Configuration Dynamic Discovery Static Discovery VWekernel Port Bindings Port Group 5 vMkernel Adapter Port Group Policy Path Status vMkernel for iSCST vMot vmki Compliant Not Used vMkernel for SCSTvMot vmk Compliant Not Used gt emos Mkernel Port Binding Details
29. ctices Frequently asked questions For more information 28 28 30 30 33 35 36 Executive summary This white paper provides detailed information on how to integrate VMware vSphere 5 0 with HP LeftHand Storage VMware vSphere is an industry leading virtualization platform and software cloud infrastructure that enables critical business applications to run with assurance and agility by virtualizing server resources Complementing this technology HP LeftHand Storage formerly known as HP P4000 LeftHand SAN solutions addresses the storage demands and cost pressures associated with server virtualization data growth and business continuity HP LeftHand Storage scales Capacity and performance linearly without incurring downtime allowing it to meet the requirements of small pay as you grow customers to the mission critical applications of an enterprise This document presents configuration guidelines best practices and answers to frequently asked questions that will help you accelerate a successful deployment of vSphere 5 0 on HP LeftHand Storage Target audience VMware and Storage administrators who are implementing vSphere 5 0 on HP LeftHand Storage New Storage feature With vSphere 5 0 VMware continues to increase both its infrastructure and application services This also influences storage deployments with VMware vSphere e VMware ESXi 5 0 converges ona thinner hypervisor architecture supporting a hardened 144 MB disk footprint
30. ding file redirection Space reclamation can only be performed with a SmartClone volume by creating a new one retaining the original small source reference with minimal delta data By removing OS dependence in separation from the User and Applications periodic SmartClone deletion and re creation helps minimize data delta growth Without these approaches SmartClone volumes may eventually occupy an entire SmartClone s volume space as delta change data SmartClone s initial value is in immediate cloning of golden image volumes Efficient space utilization objectives need to understand use and mitigation approaches to maximize success 29 Virtual machine placement and VAAI Choosing datastores for virtual machines More than one virtual machine can and generally should be stored on each VMFS datastore or volume The choice of where to put virtual machines should be driven by space considerations current VMFS datastore load the relationships of the VMs to each other and performance In general virtual machines that do not have a relationship to one another should not be mixed on the same HP LeftHand volume however there are always tradeoffs Consolidating VMs becomes mandatory as their numbers increase and larger configurations are demanded HP LeftHand features such as Snapshots Remote Copy and SmartClone volumes are very useful with virtual machines but will always affect all virtual machines on the same volume simultaneously If virtua
31. e HP LeftHand Snapshot and rollback e VMDK Affinity Virtual disks are kept together on the same datastore This is usually a best practice and enables that datastore snapshots fully contain all VM data e VM Anti affinity Two specified VMs including associated disks are placed on different datastores This is recommended when known high storage resource VMs should be separated to promote best VM SLAs HP Insight Control Storage Module for vCenter and vSphere Storage APIs for Storage Awareness VASA HP Insight Control for VMware vCenter Server is a plug in module to VMware vCenter Server which enables VMware administrators to monitor and manage their HP servers networks and storage The storage portion is handled by the storage module Administrators may clearly view the relationships between virtual machines datastores and the storage arrays and manage HP Storage directly from vCenter Instead of provisioning storage through the HP LeftHand Centralized Management Console the Storage Module of Insight Control for VMware vCenter Server supports creation of VMFS datastores and VMs from within vCenter With version 6 3 and above the HP Insight Control Storage Module HPICSM also supports VASA the HP LeftHand VASA provider is part of HPICSM and is installed through the 6 3 installation process in registering the VASA provider figure 5 The HPICSM VASA provider allows users to map the VMware objects VMs datastores RDMs to the volumes on
32. een ES source and destination is Check imbalances every f8 sid Hours I O imbalance threshold Aggressive f Conservative The I O imbalance threshold determines the amount of imbalance that Storage DRS should tolerate Aggressive setting would make Storage DRS correct small imbalanc if possible and moving it toward conservative would make Storage DRS produce recommendations only in cases when the imbalance across datastores is very high Help lt Back Next gt Cancel Ongoing balancing recommendations are made when a datastore in a Datastore Cluster exceeds user configurable space utilization or I O latency thresholds These thresholds are defined during the datastore cluster configuration figure 3 1 0 load is evaluated every eight hours by default which is also configurable When configured maximum space utilization or I O latency exceeds thresholds Storage DRS calculates all possible moves to balance VMs accordingly with a cost benefit analysis If the datastore cluster is configured for full automation VMs will move and rebalance accordingly if possible Storage DRS affinity rules enable control over which virtual disks should or should not be placed on the same datastore in a Datastore Cluster e VMDK Anti affinity Virtual disks of a VM with multiple virtual disks are placed on different datastores This helps maximize multiple datastore throughput over unique virtual disks This will generally complicat
33. for virtual machines with the two network adapters for load balancing default e The second standard virtual switch should have Two VMkernel ports network with iSCSI Port binding enabled each exclusively mapped to one network adapter as an active adapter e The third standard virtual switch should have Avirtual machine port group network for vMotion with the two network adapters for load balancing default and Avirtual machine port group network for Fault Tolerance with the two network adapters for load balancing default Figure 11 Six adapters Networking Standard Switch vSwitcho Remove Properties Virtual Machine Port Group Phisical Adapters A M Network GA vmnici 1000 Full 4 El 2 virtual machinels BA yvmnico 1000 Full 4 P000 M The a P 000 51M Vhikermel Port J Management Network ymkO 172 16 44 236 a Standard Switch vSwikchi Remove Properties WMkemel Port Fhysical Adapters J Mkernel For iSCSI B GA vmnica 1000 Full A vmkz 192 168 0 136 BA vmnicz 1000 Full 4 Vhikemel Port J vMkernel For iSCSI A wiki 192 168 0 137 Standard Switch vSwitch Remove Properties VMkemel Port Physical Adapters CJ Motion Q Ee vmnicz 1000 Full 5A yoik4 10 0 1 1 BA vmnic6 1000 Full 4 VWM kemel Port LJ Fault Tolerance a wrk 10 0 0 1 12 More than six network ports If more than six network ports are available add more ports to the iSCSI virtual switch to
34. ith HP LeftHand Application managed Snapshots raw devices may be quiesced in the context of VMs using the RDM Note that this is not supported with physical compatibility mode RDMs HP LeftHand Snapshots on VMFS datastores HP LeftHand Snapshots are very useful in a vSphere 5 environment All virtual machines stored on a single VMFS datastore i e on the same volume can be snapshot and rolled back to at any time Moreover HP LeftHand snapshots can be mounted to any vSphere 5 host without interrupting access to their source volume This may be used for data mining or testing real world data without affecting live corporate data 2 Figure 26 HP LeftHand Application Integration Solution Pack HP LeftHand OS version 9 5 required for vSphere 5 e HP P4000 Application Integration Solution Pack Home Service License HP P4000 Storage Replication Adapter SRA for Mware Application SRM Aware Snapshot Manager The HF P4000 Storage Replication Adapter SRA enables P4000 integration with the VMware Site Recovery Manager SRM to provide automated disaster recovery of VMware virtual machines OSM For MPIC Select the SRA For the VMware SRM version that is installed on your system SRA for VMware SRM 1 x 4 x SRA for VMware SRM 5 Documentation HP LeftHand Remote Copy volumes and SRM Remote Copy replicates HP LeftHand snapshots over WAN links to remote sites for disaster recovery or backup vSphere 5 environments can be p
35. itional orchestration opportunities to leverage network bandwidth automation It is also noted that 10GbE interfaces provide increased bandwidth options over 1GbE Two network ports vSphere 5 hosts with only two Gigabit 1GbE network ports are not ideal for iSCSI storage connected by the iSCSI software adapter because they compete with other network traffic and do not offer good iSCSI performance As shown in figure 9 vSphere 5 hosts with only two Gigabit network ports are configured with a single standard virtual switch that comprises both Gigabit ports and contains e Avirtual machine port group network for virtual machines with the two network adapters for load balancing default e AVMkernel port network with vMotion and management traffic with the two network adapters for load balancing default and e Two VMkernel ports with iSCSI port binding enabled on a different VLAN each exclusively mapped to one network adapter as an active adapter figure 8 Figure 9 Two network adapters Networking Refresh Standard Switch vSwikch Remove Properties Wirtual Machine Port Group Phisical Adapters 4A VM Network EA vmnicl 1000 Full 4 E 2 virtual machinets EA vmnicO 1000 Full 4 P4000 M WIN2 a P4 000 51M WM kermel Fort Ll vYMkernel For iSCSI B vmk2 192 168 0 138 VLAN ID 25 WMkemel Fort L4 vYMkernel For iSCSI A vmki 192 168 0 137 VLAN ID 25 vMkemel Port LJ Management Motion Metwaork v
36. k segment If in an HP BladeSystem environment check HP Virtual Connect for proper mapping tunneling of networks i e VLANs to uplink sets and server bays Please refer to HP Virtual Connect documentation for more information 19 Creating a new datastore on a volume After the vSphere 5 host has connected to a HP LeftHand volume the volume can be formatted as VMFS datastore or mounted as a raw device mapping RDM directly to virtual machines New VMFS datastores are formatted from within the VMware vSphere client in the new VMFS 5 format An example of the process for creating a new datastore is shown in figure 18 Figure 18 Creating anew VMFS datastore using wizards in vCenter Add Storage i a Select Disk LUN Select a LUN to create a datastore or expand the current one Name Identifier Path ID LUN Capacity Expandable or VMFS Label c Clear Name Path ID LUN Drive Type LEFTHAND iSCSI Disk naa 6000eb3 iqn 2003 10 com l 0 Non SS Properties Formatting Ready to Complete Note The HP LeftHand volume will be identified as an HP LeftHand iSCSI Disk Both HP LeftHand volumes and VMFS datastores can be expanded dynamically If space is running low on a VMFS datastore you must first expand the volume on the HP LeftHand cluster To increase the size of the HP LeftHand volume simply edit the volume in the CMC or CLI and increase its size HP LeftHand Storage immediately changes the LUN size and lays
37. l machines that have no relationship are mixed on a single volume those virtual machines will have to be snapshot rolled back remotely copied and cloned together Performance of virtual machines could also be affected if too many virtual machines are located on a single VMFS datastore The more virtual machines on a VMFS datastore the more I O and contention there is for that HP LeftHand volume particularly when multiple vSphere 5 hosts run VMs sourced on the same VMFS datastore Refer to the section on Storage Distributed Resource Scheduler Storage DRS for the new vSphere 5 feature vSphere Storage APIs for Array Integration VAAI VMware s VAAI enables HP LeftHand Storage running HP LeftHand OS version 9 5 to provide offloading from the vSphere 5 hosts for specific storage operations These functions increase the performance and efficiencies for each vSphere host and offloads vSphere host storage processing to HP LeftHand Storage where it is most efficient Originally introduced in vSphere 4 1 vSphere 5 0 further extends these features HP LeftHand Storage supports the original functional primitives at this time in HP LeftHand OS version 9 0 and 9 5 Figure 28 VAAI framework VAAI Storage vMotion VMware vSphere Provisioning VMs from template Improve provisioning disk p performance Disk array VMFS share storage pool Currently supported VAAI offloading capabilities e Full copy Enables HP LeftHand
38. load 600 E VAAI offload 400 200 Full copy Block zero Load on SAN 8000000 7455600 7000000 6000000 5000000 4000000 3000000 2000000 1000000 0 E without offload VAAI offload 3320000 620000 188000 Full copy Block zero 31 Load on Server 400 375 350 300 250 200 150 100 50 E without offload VAAI offload Full copy Block zero e Full copy clone Microsoft Windows 2008 R2 40 GB VM 7x faster and 94 percent less load e Block zero a 256 GB Virtual Machine Disk 21x faster and 92 percent less load e Hardware assisted locking 6x more VMs per volume LUN Figure 29 shows the level of improvements attributed to VAAI support with HP LeftHand Storage Best practices previously suggested that up to 16 virtual machines will function on a single volume but they might experience degraded performance depending on the hardware configuration and if they are all booted at the same time Volumes with four to eight virtual machines are less likely to affect performance With VAAI s hardware assisted locking new best practice recommendations suggest that up to 48 virtual machines will function on a single volume These figures are highly dependent upon 1 0 workload patterns as well as other variables No separate software component or setting is needed to enable VAAI VMFS datastore details will show if hardware acceleration is enabled 32 Figure 30 Manual decision tree f
39. lumes None New Schedule to Snapshot a Volume Edit Snapshot 2 GB Delete Snapshot Thin fg Convert Temp Space Snapshot Details Delete Temp Space Convert Temp Space iqn 2003 10 comlefthandnetworks lq Source Snapshot Name olume_SS_1 New Remote Snapshot Volume Name NewVolume New Schedule to Remote Snapshot a Volume Cancel Remote Snapshot Assign and Unassign Servers Roll Back Volume Edit Remote Bandwidth Help HP LeftHand Snapshots of Raw Devices HP LeftHand Snapshots of vSphere 5 RDMs are supported in exactly the same way as for physical servers which either boot from SAN or access volumes on the HP LeftHand cluster In vSphere 5 two compatibility modes for RDMs are available Virtual compatibility mode allows an RDM to act exactly like a virtual disk file including the use of VMware based snapshots Physical compatibility mode allows direct access to the volume for applications that need lower level direct raw device control RDMs offer some benefits including e Dynamic Name Resolution The identification associates each RDM with the device regardless of physical server changes adapter hardware path changes or device relocation e Distributed File Locking RDMs are available as shared raw volumes without losing data across multiple vSphere hosts accessing the same volume e Snapshots VM snapshots are possible on RDM volumes in virtual compatibility mode Combined w
40. mk 172 16 44 258 o 10 Four network ports vSphere 5 hosts with four 1GbE network ports perform better if you separate virtual machine traffic from VMkernel vMotion Management and iSCSI traffic As illustrated in figure 10 vSphere 5 hosts with four Gigabit network ports are configured with two virtual switches each comprising two Gigabit ports for load balancing on virtual machine networks management and vMotion If hosts contain two dual port network adapters use one port from each of the dual port adapters balancing high availability across both adapters For example if using two on board Gigabit adapters and a dual port Ethernet card assign port 0 from the on board adapter and port 0 from the Ethernet card to one vSphere switch and configure port 1 from the on board adapter and port 1 from the Ethernet card on a second vSphere switch This configuration provides protection from some bus or network interface card failures e The first standard virtual switch should have AVMkernel port network with vMotion and management traffic enabled with the two network adapters for load balancing default and avirtual machine port group network for virtual machines with the two network adapters for load balancing default The second standard virtual switch should have Two VMkernel ports with iSCSI port binding each exclusively mapped to one network adapter as active adapter refer to figure 8 Figure 10
41. nel ports which are connected to the iSCSI network Note that all added VMKernel ports should be listed with Port Group Policy Compliant If not check your network configuration VMkernel ports for iSCSI and their NIC teaming settings In the iSCSI Initiator Properties add the HP LeftHand cluster s virtual IP address to the Dynamic Discovery figure 13 tab and leave the port of 3260 unchanged Close the dialog and Rescan All storage adapter to look for new volumes and scan for new VMFS datastores All HP LeftHand volumes should appear in the iSCSI software adapter s devices section 15 Figure 13 Dynamic Discovery with a single Virtual IP address of a HP LeftHand Cluster iSCSI Initiator ymbba32 Properties General Network Configuration Cynamic Discovery Static Discovery Send Targets Discover S51 targets dynamically From the Following locations IP 4 host name SST Server Location dd Remove Settings Close Help ie e From the host s Configuration Storage Adapters highlight the iSCSI volumes on HP LeftHand Storage from the iSCSI software initiator s list of devices Right click and select the Manage Paths option Select the Policy Path Selection to be Round Robin VMware Select Change to apply the change in the configuration figure 14 Note that all paths list the status of Active 1 0 16 Figure 14 Manage paths and select round robin policy gt LEFTHAND iSCSI Disk
42. nnections one for each HBA initiator each with permissions to the same volumes Note that a group of hosts in the same vSphere cluster will also require authentication for each volume For easier handling a server cluster on the HP LeftHand Management Group can be created in the CMC to assign volumes to all hosts and all their HBAs at once 14 Note that most 1GbE enterprise network adapters support CRC checksum offload efficiencies and make server performance improvements negligible However 1OGbE has shown the importance of TCP processing offloads in adapters to lessen server impact Some of the 10GbE adapters have also introduced protocol convergence of 10GbE Ethernet offloading TCP BIOS boot support and iSCSI protocol support some even with multiple protocols providing new compelling reasons in leveraging these advantages Using offloading iSCSI acceleration or boot from iSCSI targets requires a tested and supported network adapter Please go to the HP Single Point of Connectivity Knowledge for tested and supported iSCSI HBAs with HP LeftHand Storage http www hp com storage spock Multi pathing for iSCSI For the iSCSI software adapter which uses multiple VMkernel ports round robin path selection for active 1 0 on all paths is optimal Native round robin iSCSI multi pathing in vSphere 5 provides superior performance and resiliency by leveraging multiple paths to HP LeftHand Storage Each path to an HP LeftHand volume will establish a
43. or determination of VM placement on HP LeftHand volumes Should another VM be put on this volume Are SAN snapshots or remote copy not being used or is it acceptable for all the other virtual machines in the datastore volume to be snapshot and restored at the same time Is the disk I O of the virtual machine that is already on the proposed datastore volume performing well Create a new datastore volume or check another one Is there enough space on the proposed datastore volume or can it be extended Put another VM on the proposed volume datastore A decision tree for VM placement is shown in figure 30 With or without VAAI virtual machine placement will impact storage 1 0 Best practices e Use at least two Gigabit network adapters on vSphere 5 hosts for improved performance and for failover of the iSCSI connections 10GbE network adapters will also increase performance e Teaming network adapters provides redundancy for networking components such as adapters cables and switches An added benefit to teaming is an increase in available 1 0 bandwidth Network bond on HP storage nodes is easily configured in the CMC by selecting two active links and enabling a bond Adaptive load balancing ALB is the most common and easiest method used on HP LeftHand storage nodes From the storage node ALB bonding supports sending and receiving for more than one iSCSI Initiator accessing the HP LeftHand storage node through bo
44. out data across the cluster accordingly without affecting the volumes availability Once the volume has been expanded you can increase the size of the VMFS datastore on one vSphere 5 host via the vSphere client Increasing the size of the VMFS datastore is preferred over adding extents on other HP LeftHand volumes to it Expanding an iSCSI LUN in the CMC 1 Goto the CMC log in to the HP LeftHand management group and locate the cluster s volumes 2 Highlight the volume to be expanded 3 Right click on the volume and click on Edit Volume 4 The Edit Volume window pops up showing the LUN size as 1 TB 20 5 Change the reported size to the desired size and click on OK figure 19 In this example it is changed to 1 2 TB Figure 19 Changing the volume size in the CMC FHP StorageWorks P4000 Centralized Management Console File Find gt Tasks Help Software upto date d Getting Started aS aj ING startle Assigned Servers Details 5 shots Remote S shot Ast Configuration Summary als napsnets EMO sNapsnors ry MG Volume A Events p HE Servers 3 Name Go 178 E may Description E Bays B Gate Cluster Cluster 3 Administration 7 d B Sites Virtual Manager Server bay 15 Cluster Performance Monitor Type Primary H Storage Systems 2 Reported Size Fe Edit Yolume A Hodel Diagnostics Data Protection Level Storage TCPAP Network A Node Provisioning Volume Name re Diagnostics Sto
45. patibility Matrix hp com go lefthandcompatibility HP Virtualization with VMware hp com go vmware HP Insight Control management software hp com go insightcontrol HP BladeSystem hp com go bladesystem HP BladeSystem interconnects hp com go bladesystem interconnects HP Systems Insight Manager HP SIM hp com go hpsim HP Virtual Connect Technology hp com go virtualconnect VMware fault tolerance demo using HP LeftHand Storage http h30431 www3 hp com fr_story 217cb39a5a037d7a3050746dc4838b02d5700d31 amp rf bm P4000 VMware VAAI Whiteboard http www youtube com watch v pkhJRwW jxXc amp To learn more about HP LeftHand Storage visit hp com go lefthand Get connected hp com go getconnected Current HP driver support and security alerts delivered directly to your desktop Copyright 2011 2012 Hewlett Packard Development Company L P The information contained herein is subject to change without notice The only warranties for HP products and services are set forth in the express warranty statements accompanying such products and services Nothing herein should be construed as constituting an additional warranty HP shall not be liable for technical or editorial errors or omissions contained herein Microsoft and Windows are U S registered trademarks of Microsoft Corporation 4AA3 6918ENW Created September 2011 Updated July 2012 Rev 1
46. phere standard switch with two physical adapters and two VMkernel ports for iSCSI Remove Properties Standard Switch vSwitchl WMkemel Fort L4 VMkernel For CSI B vmkz 192 168 0 158 WMkermel Fort L4 VMkernel for iSCSI 4 vmki 192 168 0 137 Physical Adapters e E vmnics 1000 Full 54 Lei vmnic2 1000 Full p3 Built in storage multi pathing capability uses multiple VMkernel ports for iSCSI mapped each to one network adapter which are assigned to one or multiple standard vSwitches under Configuration Networking figure 8 Figure 8 Mapping of VMkernel ports to network adapters Mkernel for iSCSI A Properties Ed Mkernel for iSCSI B Properties Ea General IP Settings Security Traffic Shaping MIC Teaming Policy Exceptions Load Balancing 7 Route based on the originating virtual port ID Network Failover Detection Link status only Notify Switches Failback Failover Order M Override switch Failover order Select active and standby adapters For this port group In a failover situation standby adapters activate in the order specified below General IP Settings Security Traffic Shaping MIC Teaming Policy Exceptions Load Balancing 7 Route based on the originating virtual port ID Network Failover Detection F Link status only Notify Switches Failback Failover Order IV Override switch Failover order Select active and standby
47. rage TCPAP Network Ha Volumes 10 and Snapshot io 178 0 Fain 17B_bay9 0 ig 500GB ext 0 io 500G _bay1 0 Eo 55 1 L_ _giss ss 1 fio Thick_on_Thin_ P4500 Palio Thick_P_4506B 0 Consumed Space ooo Target Information Description iSCSI Name Reported Size Servers 6 Thereported size shows the expansion size immediately figure 20 iSCSI Sessions Cluster Available Space 008063268 TE 21 Figure 20 The reported size has now changed and is immediately available feHP StorageWorks P4000 Centralized Management Console Pile ES Software up to date File Find Tasks Help Getting Started bast Configuration Summary T MG Hj Servers 3 t Administration B Sites Virtual Manager 4 Cluster Performance Mor H Storage Systems A Hodet re Diagnest Storage TCPAP Me A Hode re Diagnost Storage TCPAP Me m Volumes 10 ane fin 178 0 fin 1TB_bay fin 500GB e Fan 5006_bay coss 1 55 55 cii g Thick _on_ fallin Thiek P d Le n a Volume Name Description Cluster Status Server Type Reported Size Data Protection Level Consumed Space Provisioning Target Information iSCSI Name fio 178 Cluster Normal bay Primary 1 278 Network RAID 10 2 Vay Mirror 2 39 TB Created by Manual Full Created A2041 3 49 57 PM MST ign 2003 1 0 com letthandnetwarks mg 149 1tb Alarm
48. red by simply choosing the Rescan All option from the vSphere 5 host s Configuration Storage Adapters Alternatively Static Discovery may also be manually entered for each target however this is generally not considered best management practice 18 The iSCSI session status of a vSphere 5 host can be viewed in the CMC by selecting volume and selecting the iSCSI Sessions tab figure 17 All hosts and all their network adapters which are used for iSCSI should be listed with their individual iSCSI session to the selected volume Figure 17 iSCSI sessions of vSphere cluster Details Snapshots Remote Snapshots Schedules Assigned Servers iSCSI Sessions Map View gt iSCSI Sessions 4 2 Server 2 Initiator Node Name Chap Name Gateway Connection Initiator IP Port Identifier E vdi esx1 iqn 1998 01 com vmwa P4000 EMEA4 192 168 192 168 0 223 60220 230000001 E vdi esx1 iqn 1998 01 com vmwa P4000 EMEA4 192 168 192 168 0 123 55389 23d000002 r vdi esx2 iqn 1998 01 com vmwa P4000 EMEA4 192 168 192 168 0 222 53834 230000001 rE vdi esx2 iqn 1996 01 com vmwa P4Q000 EMEA4 192 1606 192 166 0 122 52210 2ad000002 Disconnecting iSCSI volumes from vSphere 5 hosts iSCSI software adapter in vSphere 5 offers options to remove iSCSI volumes from the vSphere 5 hosts It is considered best practice to gracefully 1 Unmount the VMFS datastore option is available on Configuration Storage 2 Detach the volume in the
49. rk Network cards capable of offloading processes from the CPU to the network card ASIC enhance the performance of vSphere 5 servers and balance the processing in the solution For a list of vSphere 5 supported network interfaces refer to the VMware Compatibility Guide for Network 1 0 devices http www vmware com resources compatibility and HP SPOCK Because HP LeftHand Snapshots Remote Copy volumes and HP SmartClone volumes work on a per volume basis it is best to group virtual machines on volumes based on their backup and restore relationships Virtual machines that can be backed up and restored together should share the same VFMS datastore For example a test environment made up of a domain controller and a few application servers would be a good candidate for being put on the same volume These elements could be snapshot cloned and restored as one unit HP recommends synchronizing HP LeftHand nodes and vSphere hosts with the same time server Time synchronizing your environment enhances your ability to troubleshoot problems 34 Frequently asked questions l added the HP LeftHand cluster virtual IP address to vSphere 5 host s Dynamic Discovery Send Targets list why don t I see a new target Most likely either you need to perform a Rescan All on the storage adapters or you have not correctly configured server connections and volume access in HP LeftHand Storage Also you should verify all network configurations Verify on the iS
50. rotected by Remote Copy volumes on a scheduled basis and automated by VMware Site Recovery Manager for a simple and complete disaster recovery solution HP provides an SRA for VMware SRM to integrate Remote Copy volumes seamlessly with a vSphere 5 environment For more information on Remote Copy volumes review the Remote Copy User Manual installed with the HP LeftHand Centralized Management Console vSphere 5 and VMware SRM 5 requires HP LeftHand Application Integration Solution Pack An option for SRA for VMware SRM 5 must be selected for vSphere 5 and SRM 5 support Previous version supported SRM 4 1 ina single installer option HP LeftHand OS 9 5 supports SRM 1 x 4 x or SRM 5 figure 26 A best practice is to ensure that HP LeftHand software components are always up to date Note that automated failback is now supported as part of the SRM 5 features HP LeftHand SmartClones HP LeftHand SmartClone volumes may also be used in a vSphere 5 environment Using HP LeftHand SmartClone Technology all the virtual machines stored on a single volume can be cloned instantly without replicating data Instead of a full copy of data for each SmartClone volume each SmartClone volume is merely a set of pointers back to the original volume thus consuming only a few MB of space rather than the entire size of a volume SmartClone volumes consume space for data that is written to the SmartClone volume from the time the SmartClone volume was created SmartClone volumes
51. s andDatastore Clusters IPF Ctrl L Device Drive Type Capacity f10 4TA MK SSD 106 75 GB Ctril H trl Al Storage DRS is managed in the same manner as DRS in compute resource cluster This consistently enables the smartest and best placement for new VMs and virtual disks as well as ongoing evaluation thereby enforcing or suggesting best practice load balancing based upon existing workloads Note that the configuration may be automatic or manual figure 2 Figure 2 Choosing the level of Storage DRS automation New Datastore CI E x SDRS Automation How do you want this Datastore Cluster configured Automation level No Automation Manual Mode vCenter will make migration recommendations for virtual machine storage but will not perform automatic migrations C Fully Automated Files will be migrated automatically to optimize resource usage Advanced Options sea ne c AA Clustering of datastores allows for more consistent higher level abstraction and less intimate dependency for knowledge of the HP LeftHand volumes Storage DRS thresholds govern and trigger rules for enabling expert action figure 3 Storage tiering is the process of moving data to different types of storage based upon the value of the data This tiering can be based upon performance capacity high availability or data segregation requirements By leveraging multiple clusters of HP LeftHand Storage based on different dri
52. s that the host uses an automatic path selection algorithm rotating through all available paths This implements load balancing across all of the available physical paths Load balancing spreads host 1 0 requests across all available host paths with the goal of optimizing performance storage throughput Connecting and using iSCSI volumes Enabling load balancing Virtual IP load balancing VIP LB is a setting on each defined server connection or server cluster in the CMC that allows the distribution of iSCSI sessions over the storage nodes in the HP LeftHand cluster in order to maximize performance and bandwidth utilization Most iSCSI initiators including the iSCSI software adapter in vSphere 5 and most iSCSI HBAs support this feature Virtual IP load balancing is enabled by default in the CMC and all vSphere 5 server connections on HP LeftHand Storage should have this feature enabled 17 HP LeftHand multi site deployments should use site preferencing to allow for best effort data locality This is achieved by assigning server connections to sites figure 15 Figure 15 Assigning servers to Sites in a HP LeftHand Management Group Details Map View amp Sites 2 Hame ais Site persion 2 IP Address Model Manager Special Manag RAID Status RAID Coni Servers 1 Description Controlling Se iSCSI Mode Initiator Node CHAP Name Load Bala P4000 ES 1 172 16 44 232 No CHAF required ign 1998 01 com
53. states Without this integration option snapshots are crash consistent only In flight cached data may not have been fully quiesced and may not represent an application recovery point Now with Application managed Snapshots for vSphere 5 rolling back an entire snapshot or recovering single VMs from a snapshot typically restores to point when the VMs was quiesced by a VMware snapshots The creation of this snapshot is fully integrated into the HP LeftHand Application aware Snapshot Manager To use Application managed Snapshots the HP LeftHand Application aware Snapshot Manager is installed on the vCenter Server and configured with the HP LeftHand Management Group s credentials to communicate with HP LeftHand Storage Additional configuration is performed within the HP LeftHand Centralized Management Console The server connection or all server connections in a server cluster is configured with the IP address of the vCenter Server as Controlling Server figure 23 Figure 23 Controlling Server IP Address setting in server connection Name Controlling Server IP Address 1772 16 44 232 What IP address do use SCSI Security Allow access via SOSI Initiator Hode Hame 38 01 com vmware p4000 e8x1 094d95ab How do find my initiator node name Enable load balancing information on compliant initiators Enabling load balancing on non compliant initiators can compromise volume availability To function comectly load balancing requires
54. th interfaces The best bonding option is to leverage Link Aggregation Control Protocol LACP 802 3ad if the switching infrastructure supports it From the storage node LACP bonding supports sending and receiving data from both adapters 33 On vSphere 5 hosts native storage multi pathing MPIO is using multiple VMkernel ports each mapped to a single active adapter to achieve increased bandwidth and resiliency The preferred multi path policy is round robin All paths to volumes on HP LeftHand Storage should show up as active I O VMkernel ports with iSCSI bindings should be separate from the management and virtual networks used by virtual machines If enough network adapters are available vMotion and FT should also use a separate network Separating networks by functionality iSCSI vMotion FT and virtual machines provides higher reliability and performance of those functions Hardware based flow control is recommended for HP LeftHand Storage on all network interfaces and switches The VIP load balancing feature of HP LeftHand Storage allows iSCSI sessions during volume login to be redirected to the storage cluster s least busy storage node This keeps the load on storage nodes throughout the cluster as even as possible and improves overall performance of HP LeftHand Storage This setting is enabled by default leaving it unchanged is recommended HP recommends using HP network cards that are capable of offloading TCP IP in iSCSI netwo
55. ully realize in the same manner fully realizing the benefits of scalable clustered storage architecture core offered by HP LeftHand Storage Improvements to the lock manager have also shown improvement of up to 40 percent during concurrent updates of shared resources in a vSphere cluster This impacts not only shared VMFS datastores with VMFS 5 but also when performing metadata state updates to the volume This efficiency is seen during VMFS datastore creation VM creation and deletion VM Snapshot creation and deletion vMotion and VM power on and off transition states Storage DRS Storage DRS is a new feature that provides smart VM placement across storage by making load balancing decisions which are based upon current I O performance and latency characteristics and space capacity assessments Essentially day to day operational effort is decreased either by automatically or recommending best practice implementation to configurable tolerance levels A new Datastore Cluster figure 1 aggregates a collection of datastores into a single entity Figure 1 Defining a new Datastore Cluster in VMware vCenter VCENTERS vSphere Clier File Edit View Inventory Administration Plug ins Help gt A Home gt oy Inventory gt 2 Datastores and Datastore Clusters Gara VCENTERS HP Fort Collins New Folder New Cluster New Datastore Cluster Add Host lowe Wirral Kaa hine Ctrl F lachines Hosts Datastore
56. ve technology MDL SAS SAS or SSD based and exposing volumes with different performance and capacity characteristics such as volumes from different HP LeftHand clusters with different Network RAID levels datastore clusters can simplify management for new VM placement decisions on new or current VMs including during HA movements Figure 3 Storage DRS settings for thresholds and timers f New Datastore Cl pann x SDRS Runtime Rules How do you want this Datastore Cluster configured General I O Metric Indusion SORS Automaton Select this option if you want I O metrics considered as a part of any SORS SDRS Runtime Rules recommendations or automated migrations in this datastore duster This will also enable Storage 1 O Control on all datastores in this duster IY Enable 1 0 metric for SORS recommendations 1 0 load balancing functionality is available only when all hosts connected to the datastores in this datastore duster are of version 5 0 Storage DRS Thresholds Runtime thresholds govern when storage DRS performs or recommends migrations based on your selected automation level Utilized space dictates the minimum lev sci OEA EEE Gate tin DARON tor arian par by ree deai enbia iO bity below which I O load balancing moves will not be considered Utiized Space 50 100 80 2 I O Latency Sms 100ms 15 H ms Hide Advanced Options Advanced Options No recommendations until t utilization difference betw
57. volume list from Configuration Storage Adapters on all vSphere 5 hosts and 3 Unassign the volume for all vSphere 5 hosts via the CMC If all steps above are completed the volume will no longer be displayed when rescanning the iSCSI software adapter for volumes Troubleshooting volume connectivity If new volumes are not showing up as expected try the following troubleshooting steps e Ping the virtual IP address from the iSCSI initiator to ensure basic network connectivity iSCSI HBAs typically have their own ping utilities inside the BIOS of the HBA so use the appropriate utility for your HBA vSphere 5 has a network troubleshooting utility that can be used from the Management Console or remotely through SSH to attempt a ping to HP LeftHand Storage e Ping the VMkernel port from one of the HP LeftHand storage nodes by going into the network configuration on the CMC and selecting Ping from the Network Tasks e Double check all IQNs or CHAP entries For iSCSI authentication to work correctly these must be exact e Make sure that all server connections and server clusters in the CMC have load balancing enabled If there is a mix if for example some have it enabled and some do not then hosts that do not have load balancing enabled may not correctly connect to their volumes If Pings do not work check the switch and VLAN configuration if the VMkernel ports with iSCSI binding and HP LeftHand Storage are on the same networ
58. with VMware SRM SRM does not work with volumes connected by guest iSCSI Initiators How many virtual machines should be on a volume or datastore Refer to the section Choosing datastores and volumes for virtual machines in this document Are jumbo frames supported Jumbo frames Ethernet frames configured larger than the typical 1 500 bytes i e up to 9 000 bytes are supported by HP LeftHand Storage In order for jumbo frames to be effective they must be enabled and configured end to end with the same frame size including all network adapters and switches For most Gigabit iSCSI implementations jumbo frames are not required and do not increase performance whereas 10 Gigabit iSCSI implementations may see benefits from jumbo frames for sequential workload 35 What is the maximum size iSCSI volume supported in vSphere 5 and what size VMDK The maximum volume size supported by vSphere 5 is 64 TB Note that VMFS 5 is required If upgrading ensure that VMFS datastores are running the newer VMFS 5 format The maximum VMDK size has not changed in vSphere 5 and is still 2 TB What version of HP LeftHand OS supports vSphere 5 All HP LeftHand Storage platforms running HP LeftHand OS version 9 0 and above support vSphere 5 For SRM 5 HP LeftHand OS version 9 5 is required with the new SRA 5 0 plugin For more information HP LeftHand VSA hp com go vsa HP Single Point of Connectivity Knowledge hp com storage spock HP LeftHand Com

Download Pdf Manuals

Related Search

Related Contents

Exercise 0  polymatic plus - Plasson France  CMB-679  Sivoia QS Wireless Insulating Honeycomb Shade  新電元工業株式会社 HNAV200L/H 電源モジュール 取扱説明書  Manuale d`uso MPC RotaCool, it  Samsung ProXpress M3820DW Čiernobiela laserová tlačiareň (38 ppm) Užívateľská príručka  MANUEL D`UTILISATION  有償サポート一覧はこちら  Sony 70R520A Flat Panel Television User Manual  

Copyright © All rights reserved. Failed to retrieve file