Transcript
Storage Policy-Based Management with Virtual Volume Block Storage on Hitachi Virtual Storage Platform G1000 with VMware vSphere Virtual Volumes
Tech Note By José Perez
August 2015
Feedback Hitachi Data Systems welcomes your feedback. Please share your thoughts by sending an email message to
[email protected]. To assist the routing of this message, use the paper number in the subject and the title of this white paper in the text.
Contents Product Features .................................................................................................................. 2 Hitachi Virtual Storage Platform G1000 .......................................................................................... Hitachi Storage Virtualization Operating System ............................................................................ Hitachi Compute Blade 500 ............................................................................................................ Hitachi Command Suite .................................................................................................................. Brocade Storage Area Network and Ethernet Switches................................................................. VMware vSphere 6.........................................................................................................................
2 2 2 2 2 3
Terminology Associated with VMware vSphere Virtual Volumes and SPBM ............ 4 Virtual Volume ................................................................................................................................. Storage Container ........................................................................................................................... Virtual Volume Datastore................................................................................................................. Protocol Endpoint ........................................................................................................................... Hitachi Storage Provider for VMware vCenter ................................................................................ Storage Capability Profile ............................................................................................................... Virtual Machine Storage Policies ....................................................................................................
4 4 5 5 6 6 6
Overview of a SPBM and VMware vSphere Virtual Volumes Solution on Hitachi Virtual Storage Platform G1000 ......................................................................................... 7 Storage Policy-Based Management (SPBM) .................................................................................. 7 Storage Capabilities Implemented on Hitachi Storage ................................................................... 7 Roles and Responsibilities .............................................................................................................. 9 Requirements for SPBM and VMware vSphere Virtual Volumes Implementation ........................ 10 High Level Architecture ................................................................................................................. 10
Tested Components........................................................................................................... 12 Software Components .................................................................................................................. Hardware Components ................................................................................................................. Network and Storage Infrastructure.............................................................................................. Solution Design and Use Cases....................................................................................................
12 12 14 15
SPBM Compliance and Deployment of Multi-tier Applications ................................ 19 Virtual Machine Storage Policies and Virtual Volume Datastores Compliance............................. 19 Deployment of Multi-tier Applications Using SPBM ..................................................................... 22 Default Storage Profiles ................................................................................................................ 28
Contents References ........................................................................................................................... 29 Create Storage Capabilities Profiles and Virtual Machine Storage Policies............ 30 Create Storage Containers and Capability Profiles ..................................................................... 30 Create Virtual Volume Datastores and Virtual Machine Storage Policies ..................................... 33 Provision Virtual Machines Using Storage Policies....................................................................... 36
Use PowerCLI to Perform Virtual Volume Datastores Compatibility and Virtual Machine Compliance with Storage Policies.................................................................. 38
1
Storage Policy-Based Management with Virtual Volume Block Storage on Hitachi Virtual Storage Platform G1000 with VMware vSphere Virtual Volumes Tech Note This paper shows the merits of storage policy-based management (SPBM) with VMware vSphere Virtual Volumes implemented using products from Hitachi Data Systems. There is a focus on SPBM use cases enabled by the current sets of storage capabilities exposed by Hitachi Storage Provider for VMware vCenter, based on vSphere APIs for Storage Awareness (VASA) 2.0. This solution leverages VMware vSphere Virtual Volumes on Hitachi Virtual Storage Platform G1000. Hitachi Data Systems integration with VMware vSphere Virtual Volumes enables deployment of Hitachi storage infrastructure in a software-defined, policy-controlled data center. VMware vSphere Virtual Volumes change the management, control, and awareness of virtual machine resources from the hypervisor-managed VMFS/NFS file systems into the storage array. Using Virtual Volumes, the storage array has full visibility and control over each and every individual virtual machine object (VMDK, VMX, swap) stored within the storage array. This allows using faster and more efficient SPBM and storage-offloaded features (cloning, snapshots, and so forth) at the storage array layer. Hitachi Data System focuses on providing an enterprise-level solution to empower you with rich SPBM capability control. Note — This information was developed in a lab environment. Many things affect production environments beyond prediction or duplication in a lab environment. Follow recommended practice by conducting proof-of-concept testing for acceptable results before implementing this solution in your production environment. Test the implementation in a non-production, isolated test environment that otherwise matches your production environment.
1
2
Product Features These are the products featured in this tech note.
Hitachi Virtual Storage Platform G1000 Hitachi Virtual Storage Platform G1000 provides an always-available, agile, and automated foundation that you need for a continuous infrastructure cloud. This delivers enterprise-ready software-defined storage, advanced global storage virtualization, and powerful storage. Supporting always-on operations, Virtual Storage Platform G1000 includes self-service, non-disruptive migration and active-active storage clustering for zero recovery time objectives. Automate your operations with self-optimizing, policydriven management. Virtual Storage Platform G1000 supports Oracle Real Application Clusters and VMware Metro Storage Cluster.
Hitachi Storage Virtualization Operating System Hitachi Storage Virtualization Operating System spans and integrates multiple platforms. It integrates storage system software to provide system element management and advanced storage system functions. Used across multiple platforms, Storage Virtualization Operating System includes storage virtualization, thin provisioning, storage service level controls, dynamic provisioning, and performance instrumentation. Storage Virtualization Operating System includes standards-based management software on a Hitachi Command Suite base. This provides storage configuration and control capabilities for you. This solution uses Hitachi Dynamic Tiering, a part of Server Virtualization Operating System. Separately licensed, Dynamic Tiering virtualizes and automates mobility between tiers for maximum performance and efficiency. Storage Virtualization Operating System uses Hitachi Dynamic Provisioning to provide wide striping and thin provisioning. Dynamic Provisioning provides one or more wide-striping pools across many RAID groups. Each pool has one or more dynamic provisioning virtual volumes (DP-VOLs) without initially allocating any physical space. Deploying Dynamic Provisioning avoids the routine issue of hot spots that occur on logical devices (LDEVs).
Hitachi Compute Blade 500 Hitachi Compute Blade 500 combines the high-end features with the high compute density and adaptable architecture you need to lower costs and protect investment. Safely mix a wide variety of application workloads on a highly reliable, scalable, and flexible platform. Add server management and system monitoring at no cost with Hitachi Compute Systems Manager, which can seamlessly integrate with Hitachi Command Suite in IT environments using Hitachi storage.
Hitachi Command Suite Hitachi Command Suite manages virtualized storage and server infrastructures. With usability, workflow, performance, scalability, and private cloud enablement, Hitachi Command Suite lets you build sustainable infrastructures with leading storage technologies. It helps you flexibly align with changing business requirements and maximize return on IT investments. Command Suite has expanded virtual machine management to include support for VMware vSphere Virtual Volumes.
2
3 Brocade
Storage Area Network and Ethernet Switches
Brocade and Hitachi Data Systems partner to deliver storage networking and data center solutions. These solutions reduce complexity and cost, as well as enable virtualization and cloud computing to increase business agility. This environment uses the following Brocade products:
Brocade 6520 Switch
Brocade VDX 6740 Data Center Switch
VMware vSphere 6 VMware vSphere 6 is a virtualization platform that provides a datacenter infrastructure. It features vSphere Distributed Resource Scheduler (DRS), High Availability (HA), Fault Tolerance (FT), support for VASA 2.0 and Virtual Volumes. VMware vSphere 6 has the following components:
ESXi — A hypervisor that loads directly on a physical server. It partitions one physical machine into many virtual machines that share hardware resources. vCenter Server — Management of the vSphere environment through a single user interface. With vCenter, there are features available such as vMotion, Storage vMotion, Storage Distributed Resource Scheduler, High Availability, and Fault Tolerance. Virtual Volumes — Simplifies operations through policy-driven automation that enables more agile storage consumption for virtual machines and dynamic adjustments in real time, when they are needed. It simplifies the delivery of storage service levels to individual applications by providing finer control of hardware resources and native array-based data services that can be instantiated with virtual machine granularity.
3
4
Terminology Associated with VMware vSphere Virtual Volumes and SPBM This defines terms used in this paper.
Virtual Volume A virtual volume is based on a new integration and management framework between VMware vSphere and the storage array. VMware vSphere Virtual Volumes virtualizes SAN and NAS devices by abstracting physical hardware resources into logical pools of capacity called a virtual volume (or VVol) datastore. This can be more flexibly consumed and configured to span a portion, one, or several storage arrays. VMware vSphere Virtual Volumes implements an out-of-band bi-direction control path through the vSphere APIs for Storage Awareness (VASA). It leverages unmodified standard data transport protocols for the data path, such as NFS, iSCSI, and Fibre Channel. Hitachi Storage Provider for VMware vCenter is the VASA provider for this solution. With Virtual Volumes, the virtual disk becomes the primary unit of data management at the array level. This turns the virtual datastore into a virtual machine-centric pool of capacity. It becomes possible to execute storage operations with virtual machine granularity and to provision native array-based data services to individual virtual machines. This allows you to provide the right storage service levels to each individual virtual machine.
Storage Container Storage containers are resource pool or pools of storage. A storage container provides logic abstraction for managing very large numbers of virtual volumes. As implemented by Hitachi, storage containers are one or more pools of storage similar to a dynamic provisioning pool on one of the following:
Hitachi Virtual Storage Platform family
One or more file systems (standard files systems or tier file systems) on Hitachi NAS Platform.
Use this abstraction for managing the following:
Multi-tenant environments
Various departments within a single organization
Storage containers can also set the capacity limits for a given logical grouping of virtual volumes. The setup and configuration of storage containers is by the storage administrators. They are automatically discovered in VMware vSphere when creating the virtual volume datastores. Figure 1 on page 5 shows a virtual volume-type datastore.
4
5
Figure 1
Virtual Volume Datastore A virtual volume datastore represents a storage container in VMware vCenter Server using VMware vSphere Virtual Volumes. From a virtual infrastructure administrator perspective, a virtual volume datastore is similar to any other datastore. It is used to hold virtual machines. Use virtual volume datastores with traditional VMFS and NFS datastores.
Protocol Endpoint Protocol endpoint represents an I/O access point for a virtual volume. They are not used for storage, only communication. When created, a virtual volume is not immediately accessible for I/O. To access a virtual volume, vSphere needs to issue a bind operation to a VASA provider, which creates an I/O access point for a virtual volume on a protocol endpoint chosen by a VASA provider. A single protocol endpoint can be an I/O access point for multiple virtual volumes. An Unbind operation removes this I/O access point for a given virtual volume. VMware vCenter is informed about protocol endpoints automatically through the VASA provider. Hosts discover SCSI protocol endpoints when discovering LUNs for the day. NFS mount points are automatically configured. Protocol endpoints are like LUNs or NFS mount points. They can be mounted or discovered by multiple hosts or multipathing applied. Protocol endpoints are a pass through device. In a Fibre Channel implementation, it is the sub-LUN behind the protocol endpoint that is actually storing the data. When implemented on Hitachi hardware, a protocol endpoint is called an administrative logical unit (ALU). Similar to storage containers, protocol endpoints are setup by storage administrators.
5
6 Hitachi
Storage Provider for VMware vCenter
Hitachi Storage Provider for VMware vCenter serves as out-of-band connection between VMware vSphere and underlying Hitachi storage. As the VASA provider, Storage Provider for VMware vCenter is a virtual appliance developed by Hitachi Data Systems to connect virtual machine infrastructure with storage arrays. In large IT environments, a single Storage Provider for VMware vCenter can manage multiple supported arrays. It acts as tool for VMware vSphere to manage the storage. It logically partitions storage arrays into storage containers. Effectively, a storage container replaces a logical unit number (LUN) or LDEV in the changes scheme of things. A new way of provisioning and managing storage arrays, the file system is no longer required. VASA APIs create individual virtual machine files that are known as virtual volumes, which store natively on storage containers. Data services such as snapshot, replication, and cloning are offloaded from the hypervisor to the storage array. Since the storage array is made aware of each virtual machine, perform the data services at the individual virtual machine level. Compare this to backing up or replicating an entire LUN. The result is faster operations and storage efficiency. Snapshot and replication lead time can be tremendously shortened because of granularity achieved at the storage array level.
Storage Capability Profile A storage capability profile is a pre-defined set of capabilities associated with a storage container. These capabilities are exposed to VMware vSphere using Hitachi Storage Provider for VMware vCenter in the form of rules. The rules are then used by the virtual infrastructure administrator to create virtual machine storage policies. On Hitachi storage arrays, you can do the following with these storage capabilities
Auto-generated, such as RAID type
Managed, such as performance IOPS class
Create custom capabilities, such as availability zones
Storage capabilities are defined by the storage administrators when creating the storage containers.
Virtual Machine Storage Policies Virtual machine storage policies are sets of rules used to define levels of service for the virtual machines. These policies can be used to perform placement decisions, the dynamic resource allocation of storage resources. Virtual machine storage policies are defined by the virtual infrastructure administrator.
6
7
Overview of a SPBM and VMware vSphere Virtual Volumes Solution on Hitachi Virtual Storage Platform G1000 A software-defined storage (SDS) model with VMware vSphere Virtual Volumes separates data plane from control plane of the storage.
The data plane uses protocol endpoints to serve as a gateway for direct in-band I/O between the hypervisor and Hitachi Virtual Storage Platform G1000. The control plane uses Hitachi Storage Provider for VMware vCenter, based on VASA 2.0. It serves as out-of-band connection between the hypervisor and underlying storage on Virtual Storage Platform G1000.
The control plane is the management layer responsible for controlling and monitoring storage operations. The VMware implementation of this policy-driven control plane is delivered through storage policy-based management (SPBM). The focus of this paper is using SPBM with VMware vSphere Virtual Volumes on Hitachi Virtual Storage Platform G1000.
Storage Policy-Based Management (SPBM) To enable efficient storage operations at scale, even when managing thousands of virtual machines, VMware vSphere Virtual Volumes uses storage policy-based management (SPBM). SPBM allows capturing storage service levels requirements (such as capacity, performance, and availability) in the form of logical templates (policies) to which virtual machines are associated. SPBM automates virtual machine placement by identifying available datastores that meet policy requirements and coupled with virtual volumes, it dynamically instantiates necessary data services. Through policy enforcement, SPBM also automates service level monitoring and compliance throughout the life cycle of the virtual machine. Storage capabilities profiles are advertised as part of the storage containers. Hitachi Command Suite and Hitachi Storage Provider for VMware vCenter allow adding new capability profiles, editing existing profiles, and discovering capability profiles. These capabilities are then used in storage policies to define levels of service for virtual machines.
Figure 2
Storage Capabilities Implemented on Hitachi Storage Hitachi delivers three categories of storage capabilities:
Auto-generated capabilities — This includes RAID type and encryption.
Managed storage capabilities — This includes performance class, availability class, and cost class.
Custom capabilities — This includes availability zones.
7
8 Table 1 shows more details about these storage capabilities. With the Hitachi implementation, storage administrators can discriminate the resources within the storage container and between storage containers based on performance, availability, cost, and operational recovery class attributes. Table 1. Storage Capabilities Category
Type
Capability
Value Range
Note
Managed Storage Capabilities
Performance
Performance IOPS — Class
[Tier1_IOPS/ Tier2_IOPS/ Tier3_IOPS]
Label to differentiate storage pools based on IOPS
Availability
Availability — Class
[Tier1_Latency/ Tier2_Latency/ Tier3_Latency]
Label used to differentiate storage pool based on latency
Cost
Cost — Class
[Tier1/ Tier2_/ Tier3]
Maximum cost the virtual machine administrator or tenant willing to pay for service. Used for showback and chargeback services. Storage or cloud infrastructure team determines their currency or multiple to use.
Operational Recovery*
Snapshot Backup Importance — Class
[High/ Medium/ Low]
What type of snapshot services required. Tie into Hitachi Virtual Infrastructure Integrator to manage policy.
RAID Level
[RAID10/ RAID5/ RAID6]
Values are according to storage specification
Pool Type
[HDT/ HDP]
Drive Type or Drive Speed
[SAS/ SSD/ Flash (Hitachi Accelerated Flash)/ External]
Encryption
[Yes/ No]
Snapshot
[Yes/ No]
Tag name
User definable [xx, x, x]
Autogenerated Capabilities
Custom Tag Attribute Capabilities
Custom*
Some examples that may be used for illustration:
DC Location=[,]
Availability Zone=[,]
Tenant=[,]
MaintenanceWindow=[,]
PolicyName=[,]
*These capabilities will be supported in upcoming updates of Hitachi Command Suite and Hitachi Storage Provider for VMware vCenter.
8
9 Roles
and Responsibilities
Table 2 highlights the roles and responsibilities associated with a typical implementation of VMware vSphere Virtual Volumes and SPBM on storage from Hitachi. The storage administrator controls the creation of pools, storage containers, and protocol endpoints. This person also assigns or exposes certain storage capabilities for each of the storage containers. Virtual infrastructure administrators control the creation of the virtual machine storage policies to meet their various virtual machine provisioning needs and levels of service. Table 2. Roles and Responsibilities in a Typical Hitachi Implementation of Virtual Volumes with SPBM Implementation Role
Management Interface
Responsibilities
Storage Administrator
Hitachi Command Suite
Virtual Infrastructure Administrator
VMware vCenter Server
Manage and Provision of pools
Hitachi Dynamic Provisioning
Hitachi Dynamic Tiering
Hitachi Thin Image
Manage protocol endpoints or administrative logical unit Manage storage containers Manage storage capability profiles using the Storage Containers and SPBM interface.
Manage the VMware vSphere environment.
Manage virtual machine storage policies
Hitachi Storage Provider The Storage Provider for VMware vCenter allows for VMware vCenter discovering storage capability profiles defined by the storage administrator using Hitachi Command Suite. Then, these capabilities are used in virtual machine storage policies to define levels of service for virtual machines. The only responsibility for the virtual infrastructure administrator is to register the storage arrays defined in Hitachi Command Suite in Service Provider for VMware vCenter.
9
10 Requirements
for SPBM and VMware vSphere Virtual Volumes Implementation
Table 3 provides an overview of the minimum requirements for an SPBM and VMware vSphere Virtual Volumes implementation with block storage from Hitachi. Table 3. Requirements for Virtual Volume implementation with Hitachi Virtual Storage Platform G1000 Software
Version
Hitachi Device Manager with Hitachi Command Suite
8.2.0-01
Hitachi Storage Provider for VMware vCenter
3.1.0
Hitachi Virtual Storage Platform G1000 microcode
80-03-01 or later
Hitachi Storage Virtualization Operating System Other licenses:
Local replication software
Hitachi Thin Image
Hitachi ShadowImage Heterogeneous Replication
VMware vCenter Server
6.0
VMware ESXi
6.0
For other detailed requirements and setup workflows, see the VMware vSphere 6.0 VVol Environment Deployment in Hitachi Virtual Storage Platform G1000.
High Level Architecture Figure 3 on page 11 is a logical representation of how all the components required for an SPBM and VMware vSphere Virtual Volumes implementation on Hitachi Virtual Storage Platform G1000 interact with each other.
10
11
Figure 3
11
12
Tested Components Testing of VMware vSphere Virtual Volumes and SPBM took place in the Hitachi Data Systems laboratory using Hitachi Compute Blade 500, Hitachi Virtual Storage Platform G1000, and software that is compliant from Hitachi, Ltd., Hitachi Data Systems, and VMware.
Software Components Table 4 describes the details of the software components used from Hitachi, Ltd., Hitachi Data Systems, and VMware. Table 4. Software Components Software
Version
Hitachi Command Suite, with Hitachi Device Manager
8.2.0-01
Hitachi Storage Provider for VMware vCenter
3.1.0
Hitachi Virtual Storage Platform G1000 microcode
80-03-01
Hitachi Storage Virtualization Operating System Local replication software from Hitachi and Hitachi Data Systems Hitachi Thin Image Hitachi ShadowImage VMware vCenter Server
6.0, Build 2562643
VMware ESXi
6.0, Build 2494585
Hardware Components Table 5 describes the details of the hardware components used. Table 5. Hardware Components Hardware
Detail Description
Version
Quantity
Hitachi Virtual Storage Platform G1000
Dual controllers
80-03-01-00 /00
2
64 × 8 Gb/sec Fibre Channel ports
1 TB cache memory
16 × 900 GB 10k RPM SAS disks, 2.5 inch SFF 8 × 800 GB SSD disks, 2.5 inch SFF 16 × 1.6 TB flash module drives (FMD) with Hitachi Accelerated Flash
12
13 Table 5. Hardware Components (Continued) Hardware
Detail Description
Version
Quantity
Hitachi Compute Blade 500 Chassis
SVP: A0145-H7304
1
520H A1 server blade
Brocade VDX 6740 Switches
2 Brocade 5460 Fibre Channel switch 5460: FOS 6.3.2d modules, each with 6 × 8 Gb/sec uplink ports 2 Brocade VDX 6746 Ethernet switch modules, each with 8 × 10 Gb/sec uplink ports
2 management modules
6 cooling fan modules
Half server blade 2 × 8-Core Intel Xeon E5-2680 @ 2.70 GHz
1 × 2 port 10 Gb/sec Emulex PCIe Ethernet
VDX6746: NOS 412
4 power supply modules
Brocade 6520
8-server blade chassis
BMC/EFI: 01-59/ 05-03
2
1 × 2 port 8Gb Fibre Channel Host Bus Adapter (Emulex LPe12000) 256 GB Memory 16 ×16 DIMM SAN switch with 8 Gb/sec Fibre Channel ports
2
Ethernet switch with 48 × 10 GbE ports
2
13
14 Network
and Storage Infrastructure
The network and storage infrastructure was configured based on Hitachi Data System guidelines to implement Hitachi Virtual Storage Platform G1000 with VMware vSphere. Figure 4 shows all the interconnected hardware components used in this solution.
Figure 4
14
15 Solution
Design and Use Cases
These principles guide solution design when using VMware vSphere Virtual Machines:
Storage containers support multiple storage characteristics. A storage administrator does not need to create tens of storage containers for each type of characteristic that needs to be provided. Storage containers and virtual datastores provide logic abstraction for managing very large number of virtual volumes. For example, a storage container can contain the following:
All virtual volumes created for a tenant in a multi-tenant environment
A department within single organization
Dedicated resources for certain apps
Using virtual volumes reduces the number of containers needed to the minimum.
Virtual infrastructure administrators and architects should define a best practice virtual machine storage policies based on the application needs. There is no need to create a virtual machine storage policy for every virtual machine.
A single storage container cannot span different physical storage arrays.
In order to tests some use cases around storage policies compliance and deployment of multi-tier applications using SPBM, the following setup was used:
Two Hitachi Virtual Storage Platform G1000 storage arrays were managed by Hitachi Command Suite using the same Hitachi Storage Provider for VMware vCenter instance. Three storage containers were created, each configured with one or more dynamic provisioning or dynamic tiering pools and advertising different storage capability profiles:
The “SC-GoldSilverBronze” storage container was configured with the following three dynamic provisioning pools:
The first dynamic provisioning pool containing FMD drives
The second dynamic provisioning pool containing SAS 10k RPM drives
The third dynamic provisioning pool containing SAS 7.2k RPM drives.
A pool created with Hitachi Thin Image containing SAS10k RPM drives was used for snapshot operations. Figure 5 on page 16 shows the relationship between parity groups, pools, resource groups, storage containers, and virtual volumes created on the first Hitachi Virtual Storage Platform G1000.
15
16
Figure 5
The “SC-GoldHDTSilver” storage container was configured with a dynamic provisioning pool and a dynamic tiering pool.
The dynamic tiering pool contains SSD drives and SAS 10k RPM drives
The dynamic provisioning pool contains only SAS 10k RPM drives.
A thin image pool containing SAS 10k RPM drives was used for snapshot operations.
16
17
The “SC-Bronze” storage container was configured with a single dynamic provisioning pool of SAS 7.2k RPM drives only. A thin image pool containing same drive speed was used for snapshot operations.
Figure 6 shows the relationship between system objects for these two storage containers, both configured on the second Virtual Storage Platform G1000.
Figure 6
17
18 On the VMware vCenter environment with ESXi, virtual volume datastores were created and mapped to each storage container. Then, these were and presented to all the hosts in the VMware cluster. Also, virtual volumes and virtual machine storage policies were created using the different storage capabilities from Hitachi Storage Provider for VMware vCenter, the VASA provider, for each of the storage containers. While this paper is not going to show every step how to configure every component of a solution using SPBM and virtual volumes, Create Storage Capabilities Profiles and Virtual Machine Storage Policies shows some steps related to creation of storage containers, how to define capability profiles, and how to create virtual volume datastores and virtual machine storage polices.
18
19
SPBM Compliance and Deployment of Multi-tier Applications Using SPBM and VMware vSphere Virtual Volumes to create virtual volumes on Hitachi storage arrays eliminates the need for the storage administrators to create several LUNs and VMFS datastores for virtual machines. Using virtual volumes automatic provisions the storage, in an on-demand basis during the deployment of virtual machines from VMware vCenter. Implementing virtual volumes with SPBM reduces the burden of storage administrators and empowers the virtual infrastructure Administrator to take responsibility for virtual machines and automated storage provisioning. This describes how storage policies compliance work and how virtual volume placement works when deploying multi-tier applications.
Virtual Machine Storage Policies and Virtual Volume Datastores Compliance To understand how Hitachi storage array capabilities can provide different levels of service, several virtual machine storage policies and virtual volume datastores and storage containers with different capability profiles were configured with VMware vSphere Virtual Volumes.
Table 6 on page 20 shows the capability profiles for the three storage containers with virtual volume datastores
Capabilities can be named based on the levels of performance, availability, latency, cost, RAID type, etc. For example, as shown in this table, the capability profiles are named Gold for a pool with FMD, GoldHDT for a dynamic tiering pool with SSD and SAS 10k RPM drives, Silver for a pool with SAS 10K, and Bronze for SAS 7.2.K.
Table 7 on page 20 shows the virtual machine storage policies, each with a single rule set containing multiple capabilities.
The storage admin needs to define capability profiles for each of the pools within a storage container. These capabilities dictate the storage capabilities available to vSphere.
VM storage policies can be named based on levels of performance. For example, as shown in this table, virtual machine storage policies are called Gold for Tier1/FMD, GoldHDT for HDT with SSD and SAS 10K, Silver for SAS 10K, and Bronze for SAS 7.2.K.
Table 8 on page 20 summarizes the compatibility or matching results between the virtual volume datastores listed in Table 6 versus virtual machine Storage Policies listed in Table 7.
“Create Storage Capabilities Profiles and Virtual Machine Storage Policies,” starting on page 31, describes the procedures used to create storage containers with their capability profiles, and virtual machine storage policies.
19
20 Table 6. Storage Containers and Capability Profiles Storage Containers
Pools (Dynamic Provisioning and Dynamic Tiering)
Capabilities Profile
Capabilities Performance IOPs
Availability
Cost
Drive Type and Speed
SC #1 SCGoldSilverBronze
Pool1-FMD
Post1-Gold
All
All
800
FMD
Pool1-sas10k
Pool1-Silver
Tier2_IOPS/ Tier3_IOPS
Tier2/ Tier3
400
SAS 10k RPM
Poo1-sas72k
Pool1-Bronze
Tier3-IOPS
Tier3
200
SAS 7.2k RPM
Poo2-SSDsas10k
Pool2-GoldHDT
All
Tier2/ Tier3
600
SSD, SAS 10k RPM
Pool2-sas10k
Pool2-Silver
Tier2_IOPS/ Tier3_IOPS
Tier2/ Tier3
400
SAS 10k RPM
Pool3-Bronze
Tier3_IOPS
Tier3
200
SAS 7.2k RPM
SC #2 SCGoldHDTSilver
SC #3 SC-Bronze Pool3-sas72k
Table 7. Virtual Machine Storage Policies Virtual Volume Virtual Machine (VVol VM) Storage Policies
Rule-Set or Sub-Profile 1 Performance IOPS
Availability
Drive Type/Speed
Gold
Tier1_IOPS
Tier1
FMD
GoldHDT
Tier1_IOPS
Tier2
SSD, SAS 10k RPM
Silver
Tier2_IOPS
Tier2
SAS 10k RPM
Bronze
Tier3_IOPS
Tier3
Cost (MinimumMaximum)
200 - 399
Table 8. Compatibility Results Virtual Volume Datastore
Virtual Volume Virtual Machine Storage Policies Gold
VVol-Datastore-SCGoldSilverBronze VVol-Datastore-SCGoldHDTSilver VVol-Datastore-SC-Bronze
GoldHDT
Compatible Compatible
Silver
Bronze
Compatible
Compatible
Compatible Compatible
Compliance calculation involves determining whether the virtual machine storage policy is satisfied by the virtual volume datastore current capability values defined in the capability profiles within the corresponding storage container. SPBM separates all datastores into “compatible” and “incompatible” sets based on the virtual machine requirements policies. For example, virtual machine storage policy “VVol VM Storage Policy-Gold” from Table 7 has the following requirements:
Tier1 for performance IOPS
Tier1 for availability and FMD for drive type and speed as seen on Figure 7 on page 22 (VMware vCenter).
20
21 While there are several storage containers with different capability profiles defined (Table 6 on page 20), there is only one capability profile (Pool1-Gold) from storage container “SC-Gold-SilverBronze” that matches the requirements from the virtual machine storage policy “VVol VM Storage Policy-Gold.” This capability profile supports performance IOPS (all) which includes the following:
Tier1
Availability (Tier1)
Drive type and speed (FMD).
Figure 8 on page 26 shows the compatible datastore (VVol-Datastore-SC-GoldSilverBronze) for this policy.
Figure 7
21
22
Figure 8 The virtual infrastructure administrator can use VMware vSphere Web Client to check all the compatible virtual volume datastores for certain virtual machine storage policy. However, “Use PowerCLI to Perform Virtual Volume Datastores Compatibility and Virtual Machine Compliance with Storage Policies,” starting on page 39, presents another way using PowerCLI cmdlets to check these compatibility results and virtual machine storage policies compliance. Whenever a change is made to a policy profile, the objects bound to a modified policy profile will not implicitly cause a reconfiguration to be invoked for the virtual volume objects bound to the profile. You must implicitly invoke a reconfiguration operation for that set of objects. For example, after a virtual machine storage policy has been changed, the virtual machines bound to that policy will become “non compliant” or “out of date,” and virtual infrastructure administrator should do a reconfiguration after the this change (reapply the storage policy). In the same way, if a capability profile is modified by the storage administrator, Hitachi Storage Provider for VMware vCenter automatically discovers and communicates these changes to vCenter. However, there will not be an implicit reconfiguration for the virtual machines using the affected policies/capabilities. A reconfiguration is always a follow-up task to every policy change.
Deployment of Multi-tier Applications Using SPBM The primary job of an SPBM system is to locate an acceptable virtual volume datastore when provisioning one or more objects bound to a give policy profile. Provisioning in an SPBM system allows the virtual infrastructure administrator to specify both datastore and an optional policy profile, for each individual virtual disk as well as the home object of the virtual machine. This section will continue to use the storage containers, capability profiles, virtual volume datastores and virtual machine storage policies created previously to show how SPBM automates virtual machine placement for multi-tier application virtual machines in a Hitachi implementation. Specifically, two typical multi-tier application virtual machines have been chosen: one database server and one web server. In a traditional environment, a storage administrator had to create individual LUNs for each of the VMDKs of a multi-tier application virtual machine such as a database server and web server. With virtual volumes, the storage administrator only has to create the storage container and define its storage capabilities. It is the task of the virtual infrastructure administrator to define virtual machine storage policies and apply these policies during deployment or after deployment of virtual machines.
22
23 example, a typical database server could require three virtual disks (operating system, data, and logs), and each of For these virtual disks require different level of performance. With virtual volumes, the virtual infrastructure administrator deploys the database server virtual machine from a template and applies different storage policies, as indicated in Figure 9 on page 28, where the following is true:
Operating system virtual disk uses storage policy “Silver”
Data virtual disk uses storage policy “Gold”
Logs virtual disk uses storage policy “Gold”
Another example of multi-tier application could be a web server virtual machine. This typical server could require two virtual disks (operating system and data), each virtual disk with different performance requirements. With virtual volumes, the virtual infrastructure administrator deploys the web server virtual machine and applies different storage policies as indicated in Figure 9 on page 24, where the following is true:
Operating system virtual disk uses storage policy “HDTSilver”
Data virtual disk uses storage policy “Silver”
From a virtual infrastructure administrator perspective, all the VMDKs are placed in a single virtual volume datastore. However, on the storage array, the corresponding virtual volumes are placed in different pools within the storage container, based on the storage characteristics.
23
24
Figure 9 For simplicity, Figure 9 only shows the data-virtual volumes and VMDKs.
Virtual Volume Placement Before examining placement for the virtual volume objects, know that there are five different types of virtual volume objects (config-VVol, data-VVol, swap-VVol, memory-VVol, and other-VVol). Each virtual volume maps to a different and specific virtual machine file. For every virtual machine a single virtual volume is created to replace the virtual machine directory in the system. Then there is one virtual volume for every virtual disk, one virtual volume for swap if needed, and one virtual volume per disk snapshot, and one per memory snapshot.
24
25 Analyzing the database server virtual machine with its three VMDKs, this virtual machine runs with no snapshots and the following virtual volume objects:
1 virtual volume (config-VVol) for the virtual machine home directory
1 virtual volume (data-VVol) for the operating system VMDK
1 virtual volume (data-VVol) for the data VMDK
1 virtual volume (data-VVol) for the logs VMDK
1 virtual volume (swap-VVol) for swap (while the virtual machine is running)
Figure 10 shows the database server virtual machine being deployed from a template. A storage policy “Silver” is applied for the virtual machine (home directory) and first virtual disk (operating system). The storage policy “Gold” is applied to the second virtual disk (Data) and third virtual disk (Logs). Also, the same compatible virtual volume datastore (“VVolDatastore-SC-GoldSilverBronze”) is used for all these virtual volume objects.
Figure 10
25
26 Virtual Volume Objects Seen From a Virtual Infrastructure Administrator Perspective Figure 11, from VMware vSphere Web Client, shows all these objects of the deployed virtual machine (SETO-DB01), each using different storage policies and located in the same virtual volume datastore.
Figure 11 Figure 12 shows the virtual volume objects corresponding to the same virtual machine SETO-DB01 from the VMware ESXi perspective:
The virtual machine home directory and its config-VVol
The three VMDKs with their corresponding data-VVol
The swap file with its corresponding swap-VVol
In the highlighted section, the naa.number represents the Volume ID for each of the virtual volumes at the storage array level.
Figure 12
26
27 Virtual Volume Objects Seen From a Storage Administrator Perspective Figure 13 shows the same virtual volume objects seen from Hitachi Command Suite. Here is where the virtual volume objects have been placed on the storage array:
The two data-VVols (Volume ID 29 and Volume ID 2A) corresponding to the data and logs VMDKs have been placed in a pool with FMDs which match the requirements of the virtual machine storage policy Gold. The data-VVol (Volume ID 28) corresponding to the operating system VMDK has been placed in a pool with SAS 10k RPM drives, which match the requirements of the virtual machine storage policy Silver. The config-VVol (Volume ID 27) corresponding to the virtual machine home directory has been placed in a pool with SAS 10k RPM drives, which match the requirements of the virtual machine storage policy Silver. The swap-VVol (Volume ID 2C) corresponds to the virtual machine swap.
Figure 13 Figure 14 shows the pools and capability profiles that are part of the storage container “SC-GoldSilverBronze,” which corresponds to the virtual volume datastore “VVol-Datastore-SC-GoldSilverBronze.”
Figure 14
27
28 Default
Storage Profiles
Defaults storage profiles exist for virtual volume datastores (VMware vCenter) and storage containers (Hitachi Command Suite). Default Storage Policy for Virtual Volume Datastores The storage policy “VVol No Requirements Policy” is created at the time VMware vCenter is installed. It has empty capability constraints and cannot be edited by administrators. Whenever a virtual volume datastore is created, it is assigned this profile as its default storage policy. If virtual machines and disks are provisioned on a virtual volume datastore without any storage profile, then VMware vCenter Server associates them with the “VVol No Requirements Policy” storage profile. The StorageProfile blob sent to Hitachi Storage Provider for VMware vCenter contains empty CapabilityConstraints. In this case, Hitachi Commend Suite assigns the default profile of the storage container. A virtual infrastructure administrator can assign any virtual machine storage policy as the default for the virtual volume datastore. In this case, all virtual machines created on this datastore without specifically defined storage policies use this default storage policy. Default Storage Policy for Storage Containers Default profiles are created automatically when creating a storage container. Each storage container has default profiles for each of the virtual volume types (config-VVol, data-VVol, swap-VVol, memory-VVol, and other-VVol). When deploying a virtual machine without specifically defined storage policies, the virtual volume objects for this virtual machine will be created based on the default profiles of the storage container. A storage administrator can change them as needed. See “Create Storage Capabilities Profiles and Virtual Machine Storage Policies,” starting on page 30, for instructions on how to modify default profiles for the storage containers.
28
29
References VMware vSphere Virtual Volumes:
http://www.vmware.com/products/virtual-volumes.html
http://kb.vmware.com/kb/2113013
Hitachi Data Systems deployment of vSphere Virtual Volumes:
https://www.hds.com/solutions/virtualization/vmware-vsphere/storage-solutions-for-vmware-virtualization/virtualvolumes.html
29
30
Create Storage Capabilities Profiles and Virtual Machine Storage Policies This describes how to create storage capability profiles and virtual machine storage policies.
Create Storage Containers and Capability Profiles These are defined by the storage administrator in Hitachi Command Suite.
Create Storage Containers One of the tasks for defining a storage policy-based management (SPBM) environment is the creation of a storage container. This covers only steps related to create storage containers and their capability profiles. For specific details how to create protocol endpoints (PE), pools, resource groups, and so forth, see the implementation guide from Hitachi Data Systems. Prerequisites These are the prerequisites to create a storage container.
Define a protocol endpoint (called an ALU on Hitachi storage arrays). Create the host groups and add LUN paths. Prepare a resource group, dynamic provisioning pool with Hitachi Dynamic Provisioning, and (if needed) a thin image pool with Hitachi Thin Image for the storage container must be prepared. If creating a thin image pool, then add the LDEV ID of the thin image pool volume to the resource group. When you create or delete protocol endpoints, or edit, delete, create, or expand a thin image pool, the configuration information of the storage system changes. In order to continue tasks using Hitachi Command Suite for virtual volumes after the change, you may need to refresh the storage system.
Procedure To create the storage container, do the following. 1. Connect to Hitachi Command Suite. 2. From the Administration menu, click Virtual Disk Management and then VMware. From the window that opens, click the Storage Containers and SPBM tab and then click Create Storage Container. Figure 15 shows this in Hitachi Command Suite.
Figure 15
30
31 Type the storage container name in the Name text box, and then click the values from the Storage System list and 3. the Resource Group list. A list of (undefined) capability profiles are listed for each pool included in the resource group (Figure 16).
Figure 16
Define Capability Profiles To define the capability profiles, do the following. 1. If more than one pool, select each pool and click Define Profile. The window in Figure 17 on page 32 opens. 2. Type the name for the capability profile in Name. From the Managed Storage Capabilities area, select all the capabilities to expose to VMware vCenter using Hitachi Storage Provider for VMware vCenter. 3. To create the profile, click OK. 4. Review and change, if necessary, the default profiles
31
32
Figure 17
Modify Storage Containers Default Profiles Default profiles for each of the virtual volume types are assigned automatically. However, it is possible to set these profiles manually if needed. To modify the default profiles of the storage containers, do the following. 1. On the Create Storage Containers window, click Advance Options. The window in Figure 18 on page 33 opens. 2. To change the default capabilities of a profile, select each default profile and then click Set Profile Manually.
32
33
Figure 18 While these changes are exposed to vCenter automatically by the Hitachi Storage Provider for VMware vCenter, it is possible to log on to VMware vCenter Web Client to perform a manual rescan of the storage provider, as shown in Figure 19.
Figure 19
For additional details about specific virtual volume-related workflows on Hitachi Command Suite and Hitachi Storage Provider for VMware vCenter, see the implementation guide from Hitachi Data Systems or VMware documentation.
Create Virtual Volume Datastores and Virtual Machine Storage Policies These are defined by the virtual infrastructure administrator in VMware vCenter. After the storage administrator defines storage containers with their respective capability profiles, the virtual infrastructure administrator creates virtual volume datastores and virtual machine storage policies.
33
34 Create Virtual Volume Datastores To create a virtual volume datastore, the virtual infrastructure administrator clicks the VVOL option on the New Datastore window (Figure 20). When selecting this option, a list of available storage containers will list to select the backing storage container for the virtual volume datastore.
Figure 20 The capability profiles that were defined for each storage container are now available as capability sets of the virtual volume datastore. For example, Figure 21 shows the capability sets (profiles) and default profiles of the virtual volume datastore “VVol-Datastore-SC-GoldSilverBronze.” These are the same capability profiles defined for storage container “SC-GoldSilverBronze.”
Figure 21
34
35 Create Virtual Machine Storage Polices Virtual machine storage policies can be created with rules based on tags (traditional way) or with rules based on data services. This paper focuses on storage policies with rules based on data service (Hitachi Storage Provider for VMware vCenter) only. To create virtual machine storage policies using Hitachi Storage Provider for VMware vCenter, do the following. 1. From VMware vSphere Web Client, click in Home, and then click VM Storage Policies. 2. Click the Create a new VM Storage Policy button (Figure 22).
Figure 22 3. Create a rule set with the capabilities exposed by Hitachi Storage Provider for VMware vCenter (Figure 23 on page 36).
35
36
Figure 23 4. Click Next. The list of datastores compatible with these capabilities is shown. 5. To complete the creation of the virtual machine storage policy, click Finish. The virtual machine storage policy is available to be applied during deployment or after deployment to virtual machines.
Provision Virtual Machines Using Storage Policies This is done by a virtual infrastructure administrator in VMware vCenter. Storage policies can be applied during deployment or after deployment of virtual machines. Also, as indicated in the use cases, these policies can be applied at the VMDK level. Figure 24 on page 37 shows an example where a virtual machine storage policy is applied during a clone operation.
36
37
Figure 24 For additional details about virtual machine storage policies, see the implementation guide from Hitachi or VMware documentation.
37
38
Use PowerCLI to Perform Virtual Volume Datastores Compatibility and Virtual Machine Compliance with Storage Policies There is a lot of ways to automate operations thought the VMware vSphere PowerCLI storage policy-based management cmdlets. See VMware documentation and blogs about PowerCLI and SPBM. This presents two cmdlets to check for placement compatibility and virtual machine entities compliance with storage policies. Figure 25 shows traditional cmdlets to list the storage containers and virtual machine storage policies discussed on the use cases of this paper.
Figure 25 Figure 26 shows the usage of the cmdlet Get-SpbmCompatibleStorage for checking placement compatibility, based on virtual machine storage policy requirements. The example below returns all the datastores from the CandidateStorage parameter that are in compliance with each of the virtual machine storage policies (Gold, Silver, GoldHDT and Bronze) used on the use cases of this paper. This is another way for a virtual infrastructure administrator to check compatibility of virtual volume datastores versus the virtual machine storage policies. Figure 26 on page 39 reflects the same compatibility results presented in Table 8 on page 20.
38
39
Figure 26 Figure 27 shows verifying SPBM-related configuration data of a virtual machine or virtual hard disks using the cmdlet GetSpbmEntityConfiguration. The example below shows the storage policies and compliance for each of the entities relates to the database server VM (SETO-DB01) described in the use cases of this paper.
Figure 27
39
For More Information Hitachi Data Systems Global Services offers experienced storage consultants, proven methodologies and a comprehensive services portfolio to assist you in implementing Hitachi products and solutions in your environment. For more information, see the Hitachi Data Systems Global Services website. Live and recorded product demonstrations are available for many Hitachi products. To schedule a live demonstration, contact a sales representative. To view a recorded demonstration, see the Hitachi Data Systems Corporate Resources website. Click the Product Demos tab for a list of available recorded demonstrations. Hitachi Data Systems Academy provides best-in-class training on Hitachi products, technology, solutions and certifications. Hitachi Data Systems Academy delivers on-demand web-based training (WBT), classroom-based instructor-led training (ILT) and virtual instructor-led training (vILT) courses. For more information, see the Hitachi Data Systems Services Education website. For more information about Hitachi products and services, contact your sales representative or channel partner or visit the Hitachi Data Systems website.
1
Corporate Headquarters 2845 Lafayette Street Santa Clara, CA 96050-2639 USA www.HDS.com
community.HDS.com
Regional Contact Information Americas: +1 408 970 1000 or
[email protected] Europe, Middle East and Africa: +44 (0) 1753 618000 or
[email protected] Asia Pacific: +852 3189 7900 or
[email protected]
© Hitachi Data Systems Corporation 2015. All rights reserved. HITACHI is a trademark or registered trademark of Hitachi, Ltd. ShadowImage is a trademark or registered trademark of Hitachi Data Systems Corporation. All other trademarks, service marks, and company names are properties of their respective owners. Notice: This document is for informational purposes only, and does not set forth any warranty, expressed or implied, concerning any equipment or service offered or to be offered by Hitachi Data Systems Corporation. AS-415-00, August 2015
