Introduction VSP One Block 20 Architecture

Introduction to the Hitachi Virtual Storage Platform One Block 20 Architecture
Reference Architecture
Introduction to
the Hitachi Virtual
Storage Platform
One Block 20
Architecture
CONFIDENTIAL
CONFIDENTIAL – For use by Hitachi Vantara
– For use employees
by Hitachi and otherand
Vantara employees audiences underunder
other audiences NDANDA
only.only. © Hitachi
© Hitachi Vantara
Vantara LLCAll
LLC 2024. 2024. AllReserved
Rights Rights Reserved.
1
Introduction to the Hitachi Virtual
Storage Platform One Block 20
Architecture
Table of Contents
Notices and Disclaimer ................................................................... 3
About This Document ..................................................................... 4
Introduction ..................................................................................................................................................... 4
Intended Audience .......................................................................................................................................... 4
Value Proposition ............................................................................................................................................ 4
Document Revisions ....................................................................................................................................... 4
Comments ....................................................................................................................................................... 4
VSP One Block 20 Specifications................................................... 5

Adaptive Data Reduction (ADR) ..................................................... 6
Compression Accelerator Module (CAM) ...................................... 7
Thin Image Advanced (TIA) ............................................................ 8

Front-End Configuration ................................................................. 9
VSP One Block 20 Controllers...................................................... 11
Logical Devices and MPUs ........................................................................................................................... 11
Cache Architecture ....................................................................... 12

Back-End Configuration ............................................................... 13
Dynamic Drive Protection (DDP) .................................................. 14

Summary ........................................................................................ 16
CONFIDENTIAL – For use by Hitachi Vantara employees and other audiences under NDA only. © Hitachi Vantara LLC 2024. All Rights Reserved 2
Notices and Disclaimer

© 2024 Hitachi Vantara LLC. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or
stored in a database or retrieval system for commercial purposes without the express written permission of Hitachi, Ltd., or Hitachi Vantara (collectively, “Hitachi”).
Licensee may make copies of the Materials provided that any such copy is: (i) created as an essential step in utilization of the Software as licensed and is used
in no other manner; or (ii) used for archival purposes. Licensee may not make any other copies of the Materials. "Materials" mean text, data, photographs, graphics,
audio, video and documents.
Hitachi reserves the right to make changes to this Material at any time without notice and assumes no responsibility for its use. The Materials contain the most current
information available at the time of publication.
Some of the features described in the Materials might not be currently available. Refer to the most recent product announcement for information about feature
and product availability, or contact Hitachi Vantara at https://support.HitachiVantara.com/ en_us/contact-us.html.
Notice: Hitachi products and services can be ordered only under the terms and conditions of the applicable Hitachi agreements. The use of Hitachi products is
governed by the terms of your agreements with Hitachi Vantara.
By using this software, you agree that you are responsible for:
1) Acquiring the relevant consents as may be required under local privacy laws or otherwise from authorized employees and other individuals to access
relevant data; and
2) Verifying that data continues to be held, retrieved, deleted, or otherwise processed in accordance with relevant laws.
Notice on Export Controls. The technical data and technology inherent in this Document may be subject to U.S. export control laws, including the U.S. Export
Administration Act and its associated regulations, and may be subject to export or import regulations in other countries. Reader agrees to comply strictly with all
such regulations and acknowledges that Reader has the responsibility to obtain licenses to export, re-export, or import the Document and any Compliant
Products.
EXPORT CONTROLS - Licensee will comply fully with all applicable export laws and regulations of the United States and other countries, and Licensee shall not
export, or allow the export or re-export of, the Software, API, or Materials in violation of any such laws or regulations. By downloading or using the Software, API,
or Materials, Licensee agrees to the foregoing and represents and warrants that Licensee is not located in, under the control of, or a national or resident of any
embargoed or restricted country.
Hitachi is a registered trademark of Hitachi, Ltd., In the United States and other countries.
AIX, AS/400e, DB2, Domino, DS6000, DS8000, Enterprise Storage Server, eServer, FICON, Flash Copy, IBM, Lotus, MVS, OS/390, PowerPC, RS6000,
S/390, System z9, System z10, Tivoli, z/OS, z9, z10, z13, z/VM, BCPii™ and z/VSE are registered trademarks or trademarks of International Business
Machines Corporation.
Active Directory, ActiveX, Bing, Excel, Hyper-V, Internet Explorer, the Internet Explorer logo, Microsoft, the Microsoft Corporate Logo, MS-DOS, Outlook,
PowerPoint, SharePoint, Silverlight, SmartScreen, SQL Server, Visual Basic, Visual C++, Visual Studio, Windows, the Windows logo, Windows Azure, Windows
PowerShell, Windows Server, the Windows start button, and Windows Vista are registered trademarks or trademarks of Microsoft Corporation. Microsoft
product screen shots are reprinted with permission from Microsoft Corporation.
All other trademarks, service marks, and company names in this document or web site are properties of their respective owners.
IMPORTANT: This document can only be used as Hitachi Vantara internal documentation for informational purposes only. This documentation is not meant to be
disclosed to customers or discussed without a proper non-disclosure agreement (NDA).
About This Document

Introduction
There are three models within the Hitachi Virtual Storage Platform One Block 20 (VSP One B20) family: VSP One Block 28,
26, and 24, ranging from largest to smallest. These are the newest additions to the Hitachi midsized enterprise product line,
engineered to work in an Adaptive Data Reduction (ADR) and Thin Image Advanced (TIA) environment while delivering
industry-leading performance in a small 2U, affordable form factor. The VSP One Block 20 architecture features a single, flash-
optimized Storage Virtualization Operating System (SVOS) image, running on Intel Sapphire Rapids multi-core, multi-
threaded processors, which makes it the highest-performing storage system available in a 2U form factor.
Intended Audience
This document familiarizes Hitachi Vantara sales personnel, technical support staff, approved customers, and value-added
resellers with the features and concepts of the VSP One Block 20 family. Users who will benefit the most from this document
are those who already possess an in-depth knowledge of the Hitachi Virtual Storage Platform (VSP) architecture. This
document will receive future updates to refine or expand on some discussions as the specifics of the design are better
understood or as upgrades are released.
Value Proposition
VSP One Block 20 is complemented by the Hitachi sophisticated flash-optimized cache architecture, with a focus on data
integrity, efficiency, and performance optimization in a small 2U footprint. Significant improvements in these characteristics are
delivered by the new flexible Dynamic Drive Protection (DDP) replacement for traditional RAID parity groups. Additionally,
there are improvements in Adaptive Data Reduction (ADR) performance and efficiency, which contributes to overall improved
performance; Thin Image Advanced (TIA) snapshots, which deliver more scalability and improved immutability for ransomware
protection; and expanded connectivity options with 64 Gb FC, 25GbE iSCSI, and 100GbE NVME-TCP.
Document Revisions
Revision Number Date Details
v1.0 July 2024 Initial release
Comments
Send any comments on this document to [email protected]. Include the document title, including
the revision level, and refer to specific sections and paragraphs whenever possible. All comments become the property of
Hitachi Vantara Corporation.
Thank you.
VSP One Block 20 Specifications

Hitachi Virtual Storage Platform One Block 20 (VSP One B20) is built for Adaptive Data Reduction (ADR) and Thin Image
Advanced (TIA) environments. The following table provides the specifications for each hardware component within the VSP
One Block 20 storage system models.
Table 1: VSP One Block 24, Block 26, and Block 28 Specifications
Feature VSP One B24 VSP One B26 VSP One B28
Form Factor 2U 2U 2U
Maximum 8 KiB Cache Hit

4.3 million 5.1 million 9.1 million
Read IOPS1
MPU Number of Cores 24 cores/sys 24 cores/sys 64 cores/sys

Intel Sapphire Rapids (12 cores/CPU) (12 cores/CPU) (32 cores/CPU)
2.1 GHz 2.5 GHz 2.3 GHz
Compression Accelerator Module 0 2/sys (1/CTL) 4/sys (2/CTL)
Compression Processor MPU CAM CAM
Maximum PCIe Gen 4 NVMe SSDs 24 Internal / 48 External
Maximum NVMe SSDs Raw Internal

Capacity 2.1PB (1.9PiB)
Maximum Raw External Capacity 72 PB (64 PiB) 144 PB (128 PiB) 216 PB (192 PiB)
DDR5 Cache Capacity Options 256 GiB 768 GiB 1024 GiB
Maximum Internal MPU to Cache 408 GB/s 612 GB/s 816 GB/s
Bandwidth
Maximum Number of Control Units 64 128 192

Logical Devices
16,384 32,768 49,152
Maximum Volume Size 256 TB
Maximum Host Ports 8 NVMe-TCP, 32 FC, 16 iSCSI
Data-at-Rest Encryption Available
1.
100% Random Read Cache-Hit in an ADR TIA Fibre Channel 64G environment.
Adaptive Data Reduction (ADR)

Adaptive Data Reduction (ADR) is a feature on the Hitachi storage systems that adds integrated compression and
deduplication to VSP One Block 20. ADR greatly increases the effective storage capacity of the storage system. A lossless
compression algorithm is used to reduce the number of physical bits required to represent the host-written data.
Deduplication removes redundant copies of identical data segments and replaces them with pointers to a single instance of the
data segment on disk. In VSP One Block 20, all host-facing volumes from DP Pools use ADR. Each volume in the pool has a
capacity saving attribute, with Disabled, Compression, or Deduplication and Compression (Deduplication always includes
compression) settings. The Disabled setting is only used for Hitachi Universal Replicator (UR) journals in VSP One Block 20.
Volumes in the pool with either capacity saving attribute set are referred to as Data Reduction Volumes (DRD-VOLs) or Data
Reduction Snapshot Volumes (DRS-VOLs). DRS volumes are required for Thin Image Advanced (TIA) and recommended for
most VSP One Block 20 use cases.
The data reduction engine uses a combination of inline and post-process methods to achieve capacity savings with minimum
overhead to host I/O.
Compression Accelerator Module (CAM)

Compression Accelerator Module is a hardware feature included with VSP One Block 28 and 26. CAM uses a compression
algorithm that supports 8KB and 32KB compression. CAM is not available on VSP One Block 24; therefore, the MPU is tasked
with compressing data. VSP One Block 24 uses a compression algorithm that supports 8KB compression only. Table 2 shows
the compression that was achieved during our tests using VDBench 2:1 compression and 2:1 deduplication (compratio=2
dedupratio=2).
Table 2: Compression specifications for VSP One Block 20 family storage systems
Data Reduction of Vdbench Reference Data (Compression 2:1, Deduplication 2:1)
Compression Only Deduplication + Compression
VSP E790 46% (1.9:1) 73% (3.7:1)
VSP E590 47% (1.9:1) 73% (3.7:1)
VSP One Block 24 - MPU 46% (1.9:1) 73% (3.7:1)
VSP One Block 26 - CAM 73% (3.7:1) 81% (5.3:1)
VSP One Block 28 - CAM 73% (3.7:1) 81% (5.3:1)
VSP One Block 24 is equivalent to the Hitachi Virtual Storage Platform E590 and E790 (VSP E590 and VSP E790) storage
systems in both compression and deduplication + compression, while VSP One Block 26 and 28 (utilizing the new CAM) are
better at compression and deduplication + compression than VSP One Block 24.
Thin Image Advanced (TIA)

VSP One Block 20 storage systems are designed to excel in an ADR Thin Image Advanced (TIA) environment. TIA stores
snapshots in a VSP One Block 20 storage system. If a logical data corruption occurs in the storage system because of an
incorrect data update or a virus infection, you can restore the data using the stored snapshot. As a result, ADR TIA allows
rollback capability to your data. Rollback is the process of recovering the data back to a consistent state from before a
disaster.
TIA relies on Data Reduction Snapshot (DRS) volumes. Both the Primary Volume (P-VOL) and any Secondary Volumes (S-
VOLs) must be DRS volumes.
Figure 1: VSP One Block 20 - Thin Image Advanced (TIA)

TIA is designed for application-level recovery or logical recovery, and not for hardware or physical recovery. For physical
recovery, use UR, global-active device (GAD) or a combination of UR and GAD in addition to TIA.
Front-End Configuration
Currently, the following options are provided for host connectivity:
• Fibre channel (SCSI-FCP and FC-NVMe)
• NVMe-TCP
• 25GbE iSCSI
Figure 2 shows the CHB slot pairs in the rear of the chassis. (Protocol types must be installed symmetrically between
controller 1 (lower controller) and controller 2 (upper controller)). You can use slot pair B for either CHB or DKB for a drive
expansion box.
Figure 2: VSP One Block 20 CHB locations - Rear View
Channel boards must be installed in pairs. FC CHBs support transfer rates up to 6400 MB/s, and NVMe-TCP CHBs support
transfer rates up to 12,500 MB/s. Table 3 provides additional details about the CHBs.
Table 3: Host Connectivity Options
CHB Transfer Rate CHBs Per Ports Per Ports Per Max volumes
System CHB System Per Port
32 Gb Fibre 800 / 1600 / 3200 MB/s 2/4/6/8 4 8/16/24/32 2048

Chanel
(SCSI-FCP and
FC-NVMe)
64 Gb Fibre 1600 / 3200 / 6400 2/4/6/8 4 8/16/24/32 2048

Channel MB/s
(SCSI-FCP and
FC-NVMe)
100 GbE NVMe- 12,500 MB/s 2/4 2 4/8 2048

TCP
25 GbE iSCSI 3125 MB/s 2/4/6/8 2 4/8/12/16 2048
• VSP One Block 20 FC ports operate in universal (also known as bi-directional) mode. A bi-directional port simultaneously
functions as a target (for host I/O or replication) and an initiator (for external storage virtualization or replication remote I/O
paths), with a queue depth of 1,024 for each function.
• VSP One Block 20 supports one to four pairs for Fibre Channel and can be installed in any CHB pair slot. However, to install
FC CHB options for the highest performance, start with CHB xA, and then move to the CHB xD location for two options. This
results in three FC pairs using slots CHB xA, CHB xD, and CHB xC, which provides the highest performance and allows for
the installation of a drive box extension in a future upgrade.
• VSP One Block 20 supports one or two pairs for NVMe-TCP, which must be installed in CHB pair slot A or D. 100GbE
NVMe-TCP follows a similar methodology as FC CHB; however, it can only be placed in CHB xA and CHB xD locations.
• In VSP One Block 28, when installing both NVMe-TCP CHBs and 25GbE iSCSI CHBs, at least one slot must remain
unpopulated.
• All FC ports can be used for Host IO (Target mode), remote replication, and/or virtualization.
• 10GbE iSCSI is available for remote IO and/or virtualization, but not host IO.
• 25GbE iSCSI and 100GbE NVMe-TCP can only be used for Host IO (Target mode).
• For optimal Fibre Channel or NVMe-TCP performance, use CHB pair slots A and/or D for host IO with the highest bandwidth
and performance; CHB pair slot C for replication RIO (GAD, UR, or TC), which supports change or write workloads; and I/O
pair slot B for CHB pairs. However, slot B is the only slot that can be used for DKBs to connect external drive boxes;
therefore, it must be reserved for that purpose until all other slots are used and it is determined that drive box expansion will
never be required.
VSP One Block 20 Controllers

Processing power and high-speed connectivity are at the heart of the 2U VSP One Block 20 dual-controller system, as shown
in Figure 3. The A and D channel board slots connect to the controller through 16 x PCIe Gen4 lanes and have 64 GB/s of
available bandwidth (32 GB/s send and 32 GB/s receive). The B and C slots receive eight PCIe Gen4 lanes and therefore,
have 32 GB/s of theoretical bandwidth (16 GB/s send and 16 GB/s receive).
Each controller has a multicore Intel Sapphire Rapids CPU. Both controllers are linked by two 8 x PCIe Gen4 lanes and have
64 GB/s of theoretical bandwidth (16 GB/s send and 16 GB/s receive for each PCIe connection). Like previous Hitachi
enterprise products, all VSP One Block 20 processors run a single SVOS image and share a global cache. Cache is allocated
across individual controllers for fast, efficient, and balanced memory access. The controllers are asymmetric active/active,
which allows both controllers to contribute to performance and each volume to be accessed through paths simultaneously
without a performance difference. This allows for higher peak performance than active/passive designs and simplifies path
management by having no penalty for round-robin access and not requiring preferred path administration.
Figure 3: VSP One Block 20 Controller for VSP One Block 26 and VSP One Block 28
Each CPU has eight memory channels for DDR5 memory, providing as much as 408 GB/s per CPU of theoretical memory
bandwidth and 816 GB/s per system. Finally, each controller CPU is connected through an embedded PCIe switch to the
NVMe SSDs. Each controller can establish a point-to-point connection, up to 24 drives over a 2-lane, 4 GB/s PCIe Gen4 bus.
Logical Devices and MPUs

The VSP One Block 20 controller CPUs are logically organized into multi-processing units (MPUs). There is one MPU per
controller and two MPUs per system. There are multiple uses for LDEVs in the system, some are for internal system functions
that are generally not relevant to an administrator, and others are for data volumes that are provisioned to hosts. When logical
devices (LDEVs) are provisioned, LDEV ownership is assigned (round-robin by default) to one of the two MPUs. The assigned
MPU is responsible for handling I/O commands for the logical devices it owns. The full processing power of the storage system
can be leveraged by distributing the workload across both MPUs. To leverage the full benefit of each MPU, the MPU must
have as many volumes as the number of CPU cores.
Cache Architecture
The importance of storage system cache (Dynamic Random Access Memory or DRAM) may have diminished in the era of
NVMe flash drives. However, I/O cache can still provide a significant performance boost for several reasons. First, DRAM is
faster than flash, at least by an order of magnitude. When a host requests cache-resident data (sometimes called a cache
“hit”), the command can be completed with the lowest possible latency. For example, our 8KiB random read hit testing on VSP
One Block 28 measured response times as low as 31 microseconds (0.031 milliseconds) in the ADR TIA environment. The
fastest response time observed in the 8 KiB random read miss testing was about 10x higher at 300 microseconds (0.3
milliseconds) in the ADR TIA environment. Reading data from cache is not only advantageous because DRAM is the fastest
medium, but also means that the request for data is satisfied with no additional address lookups or back-end I/O commands.
For optimizing performance, Hitachi is not content with simply caching the most recently accessed data. I/O patterns on each
volume are periodically analyzed to identify the data most likely to be accessed repeatedly. Cache is preferentially allocated to
such blocks. Meanwhile, areas of each volume identified with the lowest probability of a cache hit will not have any cache
allocated. Instead, such data is sent to the host through a transfer buffer, thereby saving the overhead of allocating a cache
segment.
Additionally, cache enhances write performance. After writes are mirrored in the DRAM of both controllers to protect against
data loss (but before data is written to flash), the host is sent a write acknowledgment. The quick response to writes allows
latency-sensitive applications to run smoothly. Newly written data are held in cache for a while to allow for related blocks to be
aggregated and written to flash together in larger chunks. The “gathering write” algorithm reduces the need for parity
operations, thereby bringing down controller and drive busy rates and improving the overall response time.
The only place in the system where the data is in a readable format by the application is in cache. When in a system using
encryption at rest, ADR, or ADR and TIA, the data on the SSD drives is in a dehydrated format and must be reconstructed in
cache using metadata. Therefore, a large quantity of cache on a modern storage system is very important and greatly helps
with performance during the rehydration process.
Back-End Configuration
VSP One Block 20 has all-NVMe back-end integrated into the controller chassis, which makes the configuration simple and
straightforward. Up to 24 NVMe SSDs can be installed in the controller chassis. Table 4 lists the supported drive types. The
NVMe SSD on VSP One Block 20 was upgraded from PCIe 3 to PCIe 4, which increased the bandwidth and performance of
each drive.
Table 4: NVMe SSD Options
Name Type Capacity
1R9RWM NVMe SSD 1.9TB
15RRWM NVMe SSD 15TB
30RRWM NVMe SSD 30TB
Dynamic Drive Protection (DDP)

With VSP One Block 20, we are introducing Dynamic Drive Protection as a replacement for Traditional RAID. In a traditional
RAID with spare drives, the spare drive remains unused until a drive failure occurs, and capacity growth is performed in large
chunks (for example, eight drives at a time). When a drive failure occurs, the spare drive tends to be a bottleneck while the
data is recreated from the failed drive to the spare drive. This is because all the data and parity must be read from the
remaining drives and written to the spare drive.
Figure 4: VSP One Block 20 - Traditional Parity Group
In Figure 4, as a characteristic of RAID 6, two parities, P and Q, are placed in each stripe, providing data protection for up to
two drive failures within a parity group.
DDP uses a similar characteristic of RAID 6 and places two parities, P and Q, in each stripe of data, providing data protection
for up to two drive failures. Figure 5 shows the stripe. However, DDP allows the group of drives within a protection domain to
exceed the parity scheme, and each data stripe only spans across the number of drives required for the parity scheme. For
example, in a traditional (6+2) RAID group, there are eight drives, and every stripe of data protection has one segment on
every drive. However, DDP groups can be larger than the protection stripe. For example, in Figure 5, there are nine drives
(and there could be more); however, each (6+2) parity stripe only needs eight drives for hardware redundancy. Therefore,
parity stripes are not all aligned to the same specific drive.
Instead of a dedicated spare, each DDP group has some spare space reserved on each drive. When a drive fails, all the
affected data stripes are rebuilt to full parity protection using the spare space. This eliminates the write bottleneck of traditional
spares. The larger the DDP group, the faster it can perform the many-drive-to-many-drive rebuild.
DDP groups can have drives added in any increment to meet customer capacity requirements. This provides a capacity that
meets customer needs without buying unwanted capacity for a complete parity group such as traditional RAID. Additionally, it
avoids stranded drive slots, which was the case with traditional RAID when there were not enough empty slots for a complete
traditional RAID group.
Figure 5: VSP One Block 20 - Dynamic Data Protection (DDP) Group

DDP has advantages such as elimination of dedicated spare drives (which does not contribute to performance) and faster
distributed rebuild times. Table 5 lists the supported DDP group data stripe parity configurations.
Table 5: Supported DDP configurations
Stripe size Minimum drives per Maximum drives per DDP

DDP group group
Entry (6D+2P) 6D+2P 9 32
Optimized (14D+2P) 14D+2P 17 32
Summary
This document provides a brief introduction to VSP One Block 20 storage systems. We are eager about the improvements in
drive protection, data reduction, snapshots, connectivity, and overall performance in a small 2U footprint of the VSP One Block
storage systems.

Introduction VSP One Block 20 Architecture

Uploaded by

Copyright:

Available Formats

Introduction VSP One Block 20 Architecture

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introduction VSP One Block 20 Architecture

Uploaded by

Copyright:

Available Formats

Introduction to the Hitachi Virtual Storage Platform One Block 20 Architecture

VSP One Block 20 Specifications................................................... 5

Thin Image Advanced (TIA) ............................................................ 8

Cache Architecture ....................................................................... 12

Dynamic Drive Protection (DDP) .................................................. 14

Notices and Disclaimer

About This Document

Revision Number Date Details

v1.0 July 2024 Initial release

VSP One Block 20 Specifications

Maximum 8 KiB Cache Hit

MPU Number of Cores 24 cores/sys 24 cores/sys 64 cores/sys

Compression Accelerator Module 0 2/sys (1/CTL) 4/sys (2/CTL)

Compression Processor MPU CAM CAM

Maximum PCIe Gen 4 NVMe SSDs 24 Internal / 48 External

Maximum NVMe SSDs Raw Internal

Maximum Number of Control Units 64 128 192

Maximum Volume Size 256 TB

Maximum Host Ports 8 NVMe-TCP, 32 FC, 16 iSCSI

Data-at-Rest Encryption Available

Adaptive Data Reduction (ADR)

Compression Accelerator Module (CAM)

Data Reduction of Vdbench Reference Data (Compression 2:1, Deduplication 2:1)

Compression Only Deduplication + Compression

VSP E790 46% (1.9:1) 73% (3.7:1)

VSP E590 47% (1.9:1) 73% (3.7:1)

VSP One Block 24 - MPU 46% (1.9:1) 73% (3.7:1)

VSP One Block 26 - CAM 73% (3.7:1) 81% (5.3:1)

VSP One Block 28 - CAM 73% (3.7:1) 81% (5.3:1)

Thin Image Advanced (TIA)

Figure 1: VSP One Block 20 - Thin Image Advanced (TIA)

Figure 2: VSP One Block 20 CHB locations - Rear View

Table 3: Host Connectivity Options

32 Gb Fibre 800 / 1600 / 3200 MB/s 2/4/6/8 4 8/16/24/32 2048

64 Gb Fibre 1600 / 3200 / 6400 2/4/6/8 4 8/16/24/32 2048

100 GbE NVMe- 12,500 MB/s 2/4 2 4/8 2048

25 GbE iSCSI 3125 MB/s 2/4/6/8 2 4/8/12/16 2048

VSP One Block 20 Controllers

Logical Devices and MPUs

Name Type Capacity

1R9RWM NVMe SSD 1.9TB

3R8RWM NVMe SSD 3.8TB

7R6RWM NVMe SSD 7.6TB

15RRWM NVMe SSD 15TB

30RRWM NVMe SSD 30TB

Dynamic Drive Protection (DDP)

Figure 4: VSP One Block 20 - Traditional Parity Group

Figure 5: VSP One Block 20 - Dynamic Data Protection (DDP) Group

Stripe size Minimum drives per Maximum drives per DDP

Entry (6D+2P) 6D+2P 9 32

Optimized (14D+2P) 14D+2P 17 32

You might also like