Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 © Violin Memory, Inc. 2014 Product Deep Dive Violin Memory 6000 Series.

Similar presentations


Presentation on theme: "1 © Violin Memory, Inc. 2014 Product Deep Dive Violin Memory 6000 Series."— Presentation transcript:

1 1 © Violin Memory, Inc Product Deep Dive Violin Memory 6000 Series

2 2 © Violin Memory, Inc MORE APPLICATIONS MORE DEVICES MORE USERS Compute Network Storage Real time, concurrent data access, heavily virtualized infrastructure Multi-Core Compute that is I/O Starved, CPU waiting for I/O Storage Must Deliver High Random IOPS & Low LatencyStorage Must Deliver High Random IOPS & Low Latency More Demand for Data, Now!

3 3 © Violin Memory, Inc  Short stroking  Wide striping  Adding SSD to legacy array  Host side read cache  “FAST”  “Easy Tier” How Do You Make Storage Go FAST? High Acquisition CostsHigh Acquisition Costs Higher Operational CostsHigher Operational Costs High Acquisition CostsHigh Acquisition Costs Higher Operational CostsHigher Operational Costs

4 4 © Violin Memory, Inc ALWAYS AVAILABLEAMAZINGLY ECONOMICALINSANELY POWERFUL Best Performance Value Lower Infrastructure Costs Eliminate IO bottlenecks Drastically reduce latency Full Redundancy Built-in Fully Hot Swappable Engineered For Flash Series Flash Array

5 5 © Violin Memory, Inc Specification VIMM Count & VIMM Capacity24x 512GiB24x 1TiB64x 512GiB64x 1TiB24x 256GiB64x 256GiB Form Factor / Flash type3U / Capacity (MLC)3U / Performance (SLC) Raw Capacity (TiB / TB)12 / 1324 / / 3564 / 706 / / 17.5 Usable Capacity 84% / 65%)6.5 / 513 / 1020 / / 313 / / 7.5 I/O Connectivity8Gb FC, 10GbE iSCSI, 40 Gb IB, PCIe G2 Maximum 4KB IOPS (Mixed)200K IOPS350K IOPS500K IOPS 750K IOPS 450K IOPS1M IOPS Maximum Bandwidth (100% Reads)1.5GB/s2GB/s4GB/s 3GB/s4GB/s Nominal Latency500 µsec (mixed)250 µsec (mixed)

6 6 © Violin Memory, Inc NO MORE “IO WAIT” 1 Million IOPS, Latency in μsec FAST BY DEFAULT No Tuning Needed Insanely Powerful. YSGet Your Storage on Moore’s Law Curve SUSTAINED EXTREME PERFORMANCE Scale without Fear

7 7 © Violin Memory, Inc Architecture Fundamentals: Violin Memory OS (vMOS)  System Operations -Web, CLI, REST  System Management -Storage virtualization -Hardware acceleration -Multi-Level Flash Optimization  Data Management -Snapshots, Clones -Thin Provisioning -Encryption -Deduplication* -Replication*

8 8 © Violin Memory, Inc vMOS – Violin Memory Operating System SYSTEM OPERATIONSSYSTEM MANAGEMENT DATA MANAGEMENT System-wide wear leveling Self-healing, integrated RAID Multi-level wide striping Die and block failure handling Efficient garbage collection LUN Management Multi-Pathing High-Availability Clustering Proactive health monitoring SNMP, CLI, UI, REST API Snapshots Clones Full-disk encryption Thin Provisioning Space management

9 9 © Violin Memory, Inc Engineered For Performance & Reliability

10 10 © Violin Memory, Inc Engineered For Performance & Reliability  Flash memory fabric -Heart of the system -4x vRAID Control Modules (VCM)  Array control modules -Fully redundant -Controls flash memory fabric -System level PCIe switching  Active/Active memory gateways -Storage virtualization -LUN configuration  IO modules -FC, 10GE, IB, PCIe Interfaces M EMORY G ATEWAYS IO M ODULES F LASH M EMORY F ABRIC 24 to 64 Hot Swappable VIMMs A RRAY C ONTROL M ODULES

11 11 © Violin Memory, Inc Multi-Level Redundancy – Hot-Swap Anything  Fans (x6)  Power Supply (x2)  VIMM (60+4 hot spares)  vRAID Controllers (x4)  Array Controllers (x2)  Memory Gateways (x2)

12 12 © Violin Memory, Inc Flash Memory Fabric – Up to 1 Million IOPS  Up to 64 Violin Intelligent Memory Modules -PCIe connected -Fully hot swappable -4 global spares  4 Active-Active vRAID Control Modules  Fabric level flash optimization -vRAID patented algorithms -Dynamic wear leveling -Multi-level Error Correction Code -Hardware based garbage collection  Performance optimization -Dynamic data wide stripping -Flash erase hiding -VIMM failure protection Flash Memory Fabric VCM Memory Gateway

13 13 © Violin Memory, Inc No SSDs ─ Violin Intelligent Memory Modules  Core building block of the Memory Fabric -256 GB SLC Flash -512 GB / 1024 GB MLC Flash -3GB to 8GB DRAM All flash metadata & write I/O buffering  Hot Swappable  Proprietary flash endurance & wear leveling extending Flash life up to 10x -Continuous data scrubbing -Advanced hardware based ECC -Automated in-place die failure handling

14 14 © Violin Memory, Inc System Level Automatic Data Placement Optimization  By default, each VCM controls 15 VIMMs -3 VIMM Protection Groups -Each comprising 5 VIMMs  Data is dynamically placed on VIMMs  Example of an incoming 4KB write -Received by MG -Forwarded to a VCM -4KB split in (4*1KB + 1 Parity) writes across 5 VIMMs in a protection group  Any VIMM failure triggers activation of a VIMM global spare and vRAID rebuild VCM VIMM VCM VIMM Protection Group P MG 4KB Write

15 15 © Violin Memory, Inc Every LUN Capable of Up to 1Million IOPS, By Default  Full system bandwidth available for every LUN  Automatic multi-level striping -Gateways to VCMs -VCM wide striping to VIMMs -VIMM wide striping on internal Flash Chips  All operations implemented in hardware, at line speed, ensuring lowest levels of latency VCM

16 16 © Violin Memory, Inc Low Level Flash Operations Can Lead to Poor I/O Latency  Read latency is low  Write is 10x to 20x longer than read  Erase is 100x longer than read Read OpsWrite OpsErase Ops SLC25µs250µs1,000µs e-MLC50µs1,500µs5,500µs MLC50µs900µs3,000µs Spike free low latency requires special handling of Erase operations

17 17 © Violin Memory, Inc Write Cliff Affects All Flash Solutions To Some Degree  New Write operations get queued behind Erase operations  Up to 60% performance drop  Real issue is that Erase operations also get in the way of Read operations  Mitigating or eliminating the Write Cliff requires special flash management logic Transient Random Write Bandwidth Degradation Source: Nersc “Write Cliff”

18 18 © Violin Memory, Inc Patented Algorithms Deliver Spike Free Low Latency  Background garbage collection ensures free pages for all incoming writes  Garbage collection implemented in hardware within each VIMM for line rate performance  Garbage collection tightly scheduled & orchestrated at the system level to not affect system performance  Garbage collection allowed one VIMM per Protection Group at a time VCM

19 19 © Violin Memory, Inc Patented Algorithms Deliver Spike Free Low Latency  Background garbage collection ensures free pages for all incoming writes  Garbage collection implemented in hardware within each VIMM for line rate performance  Garbage collection tightly scheduled & orchestrated at the system level to not affect system performance  Garbage collection allowed one VIMM per Protection Group at a time VCM

20 20 © Violin Memory, Inc vRAID Erase Hiding In Action  Reads never blocked by garbage collection (vRAID rebuild on remaining 4 VIMMs)  System level orchestration enables sustained low latency for mixed workloads VCM P 4KB Read vRAID Rebuild

21 21 © Violin Memory, Inc World Record Breaking Performance  June 29, TPC-E World Record  May 9, TPC-C World Record  May 23, TPC-C World Record  June 22, 2011 – File System World Record  December 8, TPC-C World Record  September 12, 2012 – VMmark 2.1 World Record  September 18, 2012 – VMmark 2.1 World Record  September 27, 2012 – TPC-C World Record  October 02, 2012 – VMmark 2.1 World Record  November 13, 2012 – VMmark 2.1 World Records (5 of them)

22 22 © Violin Memory, Inc R EDUCE S TORAGE C OSTS BY 7 X C OMPARED TO D ISK Never Overprovision U NMATCHED O PERATIONAL C OST Plug and play experience N EAR I NSTANT ROI Optimize Server and License Costs Amazingly Economical. Reduce Cost Across Your Infrastructure

23 23 © Violin Memory, Inc Storage Cost Per Application Is What Matters Database Requirements 1TB 20K IOPS & Tier 1 Disk Array - High Performance HDD - $4/Raw GB IOPS per disk - 146GB per disk Violin Memory Flash Memory - $5/Raw GB | $8.5/Usable GB - vRAID - 750k IOPS for any size LUN 1TB 750K IOPS 100 Disks * 146 = 14.6 Raw TB 20K IOPS

24 24 © Violin Memory, Inc Application Owners Pay 7x Less on Violin Database Requirements 1TB 20K IOPS & Violin Memory 6264 Tier 1 Disk Array Raw TB - $4/Raw GB - 1 Usable TB - $8.5 / Usable GB $58,400 For This Database $8,500 For This Database Application Storage Costs is 7x Lower With Violin!

25 25 © Violin Memory, Inc Simple Operations  Provision storage and Go! -Select LUN capacity and let vRAID automate placement -No tuning required -Hot swap for non disruptive operations  Seamlessly handle performance spikes -Customer example: Rogue full table scans in dba scripts System handled the load spikes and still met core application SLAs  Advanced Graphical User Interface -Fully customizable dashboard -Detailed performance statistics -Supported as a vCenter Plug-In

26 Violin Memory Inc. Proprietary 26

27 Violin Memory Inc. Proprietary 27

28 28 © Violin Memory, Inc Violin Symphony: Manage PB’s in a Flash! Manage 100’s of Violin flash arrays through a single interface Enable multi-tenancy with role based access control and Smart Groups Share information through custom reports with up to 2 years of historic data Achieve pro-active wellness with advanced health & SLA monitoring Personalize visibility through fully customizable dashboards and gadgets

29 29 © Violin Memory, Inc Eliminate “I/O Wait”; Reduce HW & SW Costs the Speed of Memory More Ops/Sec With Less CPU Cores More Ops/Sec with Less DRAM Cache Less Software Licenses I/O Wait CPU Cycle with Magnetic Disk: 80% Wait 20% Work t CPU Cycle with Memory Storage: 5% Wait 95% Work t

30 30 © Violin Memory, Inc VMworld 1 Million IOPS – 2011 vs Engines, 960 drives 1 Million Read IOPS 5 Racks or 210RU – 32,000 Watts 2 Violin 6616 Memory Arrays 1 VM at 1 Million IOPS (Random R/W Mix) 6 RU (97% less) – 3,600 Watts (90% less)

31 31 © Violin Memory, Inc Bringing the speed of Flash to all your applications T ARGETING A LL A PPLICATIONS  Enterprise Applications on Legacy SAN  Enterprise Applications on Memory SAN  Scale Out Applications D ISRUPTIVE E CONOMICS  Reduced Capex  Streamlined Opex  Ready for Petabyte Scale

32 32 © Violin Memory, Inc Back-up slides

33 33 © Violin Memory, Inc Violin 6264: A New Standard in Performance Economics HIGHER EFFICIENCY BETTER ECONOMICS SAME FOOTPRINT DOUBLE CAPACITY Violin 6232Violin X 3X

34 34 © Violin Memory, Inc Violin 6264 Flash Memory Array at a Glance 50% Lower Power 750K IOPS (Peak 70:30) Memory Disk $/GB 19nm Process Geometry 64 TiBs of Capacity in the Same 3U Form Factor

35 35 © Violin Memory, Inc Comparing 6264 and 6232 Hardware & Software  6264 requires 250W less than W for 6264 versus 1750W for Result of more power efficient VIMM hardware design  6264 specific hardware improvements -New 1TiB MLC VIMMs -New chassis with better cable management -FC, iSCSI and IB configurations come with a new ACM Internal clustering for vMOS 6 40GE native port – for future use, enabling data movement across arrays -PCIe configuration leverages same ACM as 6232  6264 requires Array Firmware 6.2 and above -Memory Gateway software is equivalent functionality to vMOS 6.0 -Memory Array firmware adds resilience and support for new ACMs and VIMMs  vMOS 6.3 will support all 6600 and 6200 Series arrays

36 36 © Violin Memory, Inc Array Control Module with 40Gbps Ports

37 37 © Violin Memory, Inc Back panel view – 6264 FC/iSCSI/IB – New ACM


Download ppt "1 © Violin Memory, Inc. 2014 Product Deep Dive Violin Memory 6000 Series."

Similar presentations


Ads by Google