Virtual Storage Platform

Hitachi Virtual Storage Platform, also known as VSP was first introduced in September, 2010.[1] This storage platform builds on the design of Universal Storage Platform V, originally released in 2007.[2]

At the heart of the system is the HiStar E-Network, a network crossbar switch matrix. This storage platform is made up of different technologies than USP and USP V. The connectivity to back-end disks is via 6Gbit/s SAS links instead of 4Gbit/s Fibre Channel loop. The internal processors are now Intel multi-core processors, and in addition to 3.5-inch drives support has been added for 2.5 inch small-form factor HDDs. The VSP supports SSD, SAS and SATA drives.[3]

Features included:[4]

• The ability for growth in three ways:
  • Scale up to meet increasing demands by dynamically adding processors, connectivity and capacity in a single unit. This enables tuning the configuration for optimal performance for both open systems and mainframe environments.
  • Scale out to meet demands by dynamically combining multiple units into a single logical system with shared resources. Support increased needs in virtualized server environments and ensure safe multitenancy and quality of service through partitioning of cache and ports.
  • Scale deep to extend the functions of Hitachi Virtual Storage Platform to multivendor storage through virtualization. Offload less-critical data to external tiers in order to optimize the availability of the tier one resources.
• Supports automated storage tiering, known as Dynamic Tiering, to automate the movement of data between tiers to optimize performance.
• Front to back cooling airflow for more efficient cooling
• Improved capacity per square foot and lower power consumption compared to the USP V.
• Enables virtualization of external SAN storage from Hitachi and other vendors into one pool
• Supports online local and distance replication and migration of data nondisruptively internally and between heterogeneous storage, without interrupting application I/O through use of products such as Tiered Storage Manager, ShadowImage, TrueCopy and Universal Replicator.
• Single image global cache accessible across all virtual storage directors for maximum performance.[5]
• Automated wide-striping of data, which allows pool balancing and lets volume grow or shrink dynamically.[5]
• The system can scale between one and six 19-inch rack cabinets. It can hold a maximum of 2,048 SAS high-density 2.5-inch drives for 1.2 petabytes of capacity, or 1,280 3.5-inch SATA drives for a maximum capacity of 2.5 petabytes.[6]
• Supports thin provisioning and storage reclamation on internal and external virtual storage
• Provides encryption, WORM and data shredding services, data resilience and business continuity services and content management services.

Delivering on enterprise demands for real-time customer engagement requires more than a fast storage array. It requires an end-to-end approach to data management that leverages both the storage operating system and flash media, to deliver low latency performance even as data levels grow exponentially. Current all-flash arrays (AFAs) rely on performance management to be handled in the array controller along with all other operations, such as data reduction. As data levels increase, they can cause controller bottlenecks, sporadic response times and a poor customer experience. To offset this issue, IT organizations have been forced to make tradeoffs in the number of workloads or amount of data they store on an individual AFA. These decisions result in the need to deploy and support more systems, raising costs and increasing management complexity. Hitachi understands this and has enhanced Hitachi Storage Virtualization Operating System RF (SVOS RF), which powers our award-winning Hitachi Virtual Storage Platforms (VSP). Storage Virtualization Operating System integrated with Hitachi Accelerated Flash (HAF) fundamentally changes this paradigm. Now organizations can engage customers faster, simplify storage operations and leverage the cloud for a superior return on investment. HAF is powered with flash optimizations to SVOS RF and unique solid-state hardware design. This approach eliminates these performance tradeoffs and answers demands for intelligent high performance, predictable, submillisecond response time and improved data center efficiency. With more than 350 flash patents, SVOS RF optimizations are engineered to accelerate the I/O path for access to flash devices. The result is a complete “flash-accelerated” refresh of the operating system that delivers significantly improved I/O processing, higher multithreading support and faster internal data movement. It also reduces response times considerably.

With SVOS RF, organizations benefit:

• Flash-aware I/O stack accelerates data access.
• Leading storage virtualization consolidates investments.
• Best-in-class business continuity prevents outages.
• Adaptive data reduction services reduce storage needs.
• Direct connect to cloud assures predictable, ongoing IT costs.

FMD disk technology is the fastest disk technology before NVMe technology came along. NVMe technology provides a much faster and reliable experience. In addition, it is predicted that 75% of the disk technology in the world will be to NVMe by 2022.

Running on Hitachi VSP family systems, HAF enables sub-millisecond delivery on a petabyte scale. The purpose of this white paper is to take a close look at Hitachi Accelerated Flash. The discussion focuses on the uniqueness of this solution. It considers why and how these technologies can meet the increasing IT challenges of large and small enterprises as they focus on the management challenges of their high-velocity data.

Virtual Storage Platform specifications in 2010 were:[7][8]

• Frames (19-inch racks) - Integrated Control Chassis/Disk Chassis Frame (2) and up to 4 optional Disk Chassis Frames
• HiStar-E Network - Number of grid switches 4 pair (8)
• Aggregate bandwidth (GB/sec) - 192
• Aggregate IOPS - 5,600,000
• Cache Memory
  • Number of data cache adapters (DCA) 2-16
  • Module capacity 2-8GB
  • Maximum cache memory 1,024GB
• Control Memory
  • Number of control memory modules 2-8
  • Module capacity 2-4GB
  • Maximum control memory 32GB
• Front End Directors (Connectivity)
  • Number of Directors 2-24
  • Fibre Channel host ports per Director - 8 or 16
  • Fibre Channel port performance - 2, 4, 8 Gbit/s
  • Maximum Fibre Channel host ports - 192
  • Virtual host ports - 1,024 per physical port
  • Maximum IBM FICON host ports - 192
  • Maximum IBM FCoE host ports - 96
• Logical Devices (LUNs) — Maximum Supported
  • Open systems 65,536
  • IBM z/OS 65,536
• Disks
  • Type: Flash 200GB (2.5"), 400GB (3.5")
  • Type: SAS 146, 300, 600GB (2.5")
  • Type: SATA II 2TB
  • Number of disks per system (max) 2.5" - 2,048; 3.5" - 1,280
  • Number spare disks per system (min-max) 1-256
  • Maximum Internal Raw Capacity - (2TB disks) 2.52PB
  • RAID 1, 5, 6 support
  • Maximum internal and external capacity 255PB
  • Max. Usable Internal capacity RAID-5 (7D+1P)
    • OPEN-V 2,080.8TB
    • z/OS 3390M 2,192.2TB
  • Max. Usable Internal Capacity RAID-6 (6D+2P)
    • OPEN-V 1,879TB
    • z/OS 3390M 1,779.7TB
  • Max. Usable Internal Capacity RAID-1+0 (2D+2D)
    • OPEN-V 1,256.6TB
    • z/OS 3390M 1,190.2TB
• Virtual Storage Machines 32 max
• Back End Directors 2-8
• Operating System Support
  • Mainframe
    • IBM: z/OS, z/OS.e, OS/390, z/VM, VM/ESA, zVSE, VSE/ESA, MVS/XA, MVS/ESA, TPF, Linux for IBM S/390 and zSeries;
  • Open systems
    • HP: HP-UX, Tru64 UNIX, Open VMS
    • IBM: AIX
    • Microsoft: Windows Server 2000, 2003, 2008
    • Novell: NetWare, SUSE Linux
    • Red Hat: Enterprise Linux
    • Oracle: Solaris
    • VMware: ESX Server

Hitachi Command Suite (formerly Hitachi Storage Command Suite) delivers management along three dimensions in support of the system as it does for all of the company's systems. 3D management simplifies operations and lowers costs along three distinct dimensions:

• Manage up to scale with the largest infrastructure deployments
• Manage out with breadth to manage storage, servers and the IT infrastructure
• Manage deep with the integration required to manage the multivendor resources of today’s complex data centers

Command Suite provides integrated storage resource management, tiered storage and business continuity software solutions allowing customers to align their storage with application requirements based upon metrics including Quality-of-Service, Service Level Objectives, Recovery Time Objectives and Recovery Point Objectives.[9] Hitachi Command Suite employs a use case-driven, step-by-step wizard-based approach that allows administrators to perform tasks such as new volume provisioning, configuration of external storage, and creation/expansion of storage pools easily on the fly.[10]

Hitachi Command Suite is composed of the following software products:[11]

• Hitachi Basic Operating System[12]
  • Hitachi Dynamic Provisioning
  • Hitachi Device Manager
  • Hitachi Dynamic Link Manager Advanced
• Hitachi Basic Operating System V[13]
  • Hitachi Universal Volume Manager
• Hitachi Dynamic Tiering
• Hitachi Command Director
• Hitachi Storage Capacity Reporter, powered by APTARE
• Hitachi Tiered Storage Manager
• Hitachi Tuning Manager
• Hitachi Virtual Server Reporter, powered by APTARE

Hitachi Command Suite also supports management interfaces such as SNMP and SMI-S.[14]