EMC unleashes the power of flash with next-generation VNX storage platforms

The next-generation VNX (“VNX2″) is designed from the ground up with Flash in mind and the goal of delivering the lowest latency, highest performance, and lowest cost per IO and per GB, of any mid-range storage platform.

Highlights include:

  • ŸSandy Bridge CPUs and PCIe Gen 3 – 40GbE and 16Gb FC ready
  • Multi-Core Optimisation (MCx) for RAID, Cache and FAST Cache
  • ŸIncreased cores, memory, drive count and FAST Cache capacity
  • Scaling to one million 8k read IOPS
  • ŸSupport for 100GB, 200GB and 400GB lower-cost eMLC Flash drives
  • De-duplication of block LUNs included as standard
  • Adaptive Cache management with a shared pool for reads and writes
  • Improved FAST Cache performance and warm-up times
  • Active/Active LUN access for improved performance and availability
  • ŸMore granular FAST VP segment sizes – down from 1GB to 256MB
  • Permanent drive sparing and drive portability
  • ŸVNX Monitoring and Reporting included as standard
  • Automated online file system migrations from Celerra and VNX systems

EMC_VNX_NG1

Compared to the first generation VNX scaling is as follows:

  • VNX5400 vs. VNX5300 (2 x performance and 2 x drive count)
  • VNX5600 vs. VNX5500 (2 x performance and 2 x drive count)
  • VNX5800 vs. VNX5700 (2.5 x performance and 1.5 x drive count)
  • VNX7600 vs. VNX7500 (3 x performance)
  • VNX8000 vs. VNX7500 (5 x performance and 1.5 x drive count)

We expect the new VNX to be deployed with significantly more Flash than existing mid-range storage platforms as it has the power to handle far more IOs and the use of eMLC (Enterprise Multi Level Cell) drives will significantly lower the cost per GB – when combined with de-duplication the cost lowers even further.

SLC (Single Level Cell) drives are rated at approximately 5,000 IOs per second and eMLC at 3,500, this compares to 140 for 10K and 90 for 7.2K drives – so it is easy to see why Flash is so game changing. FAST Cache will continue to use SLC drives, but data volumes are expected to move to eMLC.

But what about reliability?

Each SLC cell can be re-written 30 times a day for 5 years and this drops to 10 times a day for eMLC. This may not sound a lot, but a 100GB SLC drive has 12.5 million 8K cells and wear levelling spreads the writes across all cells, so this equates to 4,300 writes per second per drive non-stop for five years.

For a 400GB eMLC drive this equates to 6,000 writes per second – so reliability for SLC and even eMLC drives should be of no concern, and in fact they should be more reliable than traditional drives.

If you want to learn more about the next generation EMC VNX then attend one of our technical seminars (12th September at Pinewood Studios or 15th October at EMC’s City Office in London) – you can find out more and register at http://www.snsltd.co.uk/emc-vnx-futures-technical-briefing/

Mark Burgess has worked in IT since 1984, starting as a programmer on DEC VAX systems, then moving into PC software development using Clipper and FoxPro. From here he moved into network administration using Novell NetWare, which kicked-off his interest in storage. In 1999 he co-founded SNS, a consultancy firm initially focused on Novell technologies, but overtime Virtualisation and Storage. Mark writes a popular blog and is a frequent contributor to Twitter and other popular Virtualisation and Storage blog sites.
twittergoogle_pluslinkedinmailtwittergoogle_pluslinkedinmailby feather

About Mark Burgess

Mark Burgess has worked in IT since 1984, starting as a programmer on DEC VAX systems, then moving into PC software development using Clipper and FoxPro. From here he moved into network administration using Novell NetWare, which kicked-off his interest in storage. In 1999 he co-founded SNS, a consultancy firm initially focused on Novell technologies, but overtime Virtualisation and Storage. Mark writes a popular blog and is a frequent contributor to Twitter and other popular Virtualisation and Storage blog sites.

Leave a Reply