HP Storage Tech Day

I was recently (two weeks ago) lucky enough to be invited to attend an HP Storage Tech Day.  It was a great opportunity to learn some great tech and network with some great people. I had every intention of getting this post out earlier (honestly!), but the day
job and other engagements have kept me beyond busy.  At least I got it out before PEX!

Full disclosures: I work for a VAR where we sell and implement both HP and EMC storage solutions. I should also mention that this was an all-expenses paid even.  Oh, and we got a really cool jacket with our twitter handle on the back (which should make me more identifiable at PEX).

The first day started off with an introduction by each of the attendees.  After that, Craig Nunes got up to give us what was supposed to be the only Marketing part of the day.  He shared with us some charts that showed the explosive growth HP has seen in the last year, especially around the 3PAR and backup appliance product lines. He also introduced the HP Storage product line:

  • P10000 (3PAR) – Utility platform for ITaaS and Cloud
  • P4000 (LeftHand) – Scale out and virtualization
  • X#000 (IBRIX) – Scale out NAS and big data
  • B#000 (StoreOnce) – Deduplication and disk-based backup


Brad Katz then gave us an overview of the P4000 line, concentrating on the following:

  • VSA – a fully functional virtual appliance version of the SAN/IQ operating system that runs on the P4000 appliances.  It can even be brought into a normal P4000 cluster and clustered across hosts.  It’s far more feature rich and scalable than VMware’s vSphere Storage Appliance.
  • Metrocluster – this is the capability to geographically separate nodes within a P4000 cluster to create an active-active array across multiple sites.  A very cool technology.  I have customers that run this capability in their Production environments, and it works great.
  • P4800 – by far the coolest technology in the P4000 family.  This solution consists of two BL460 blades that run the VSA and are SAS connected to an MDS disk shelf.  The disk shelf is divided into two drawers (one for each blade) that pull out and contain up to 70TB raw hot swappable drives.  An extremely dense storage device that utilizes the entire depth of the rack instead of just the front panel.  Check this one in person if you ever get the chance.

There is also a cool new deployment tool we were able to play with to deploy the VSA very simply and quickly.  They call it “zero to VSA”.

Being an iSCSI-only array, some quickly dismiss this product line for enterprise storage, but I have many happy customers running P4000 storage.  Plus I know some NDA futures of this product line, and there are a lot of cool things to come this year.  Some of which should definitely change this perception.


Next, Steve Johnson gave us a briefing of the StoreOnce products, concentrating on the top of the line B6200.  Interesting factoid: HP’s first backup customer was IBM, who OEM’d HP tapes in the 80’s.

The B6200 is a combination of DL380s and P2000 hardware and StoreOnce and IBRIX software.  It can provide both NAS and FC connectivity.  It also provides deduplication using the StoreOnce engine, which is utilized across all the HP products offering deduplication, a 4k block size and sparse indexing.  It can scale out to 512TB usable storage and max throughput of 28TB/hr.

Architecturally, it looks a lot like a general purpose SAN and contains all the redundancy you’d expect (including utilizing RAID6 (4+2).  One major downside: replication is an add-on license.  I find this a major gap since this array has a strong story as a destination target for multiple StoreOnce devices in multiple locations.


James “JR” Richardson provided us the 3PAR presentation.  What a great speaker.  Very engaging, new his stuff well (he was one of the first 3PAR employees) and passion to spare.  I have every intention of getting him to Omaha someday soon for a lunch and learn and/or a series of customer meetings.

He spent most of the time laying out the three main differentiators of the 3PAR line: Multi-tenant, Efficient & Autonomic

  • Multi-tenant
    For a 3PAR array “multi-tenant” is more than having multiple customers in a single array.  It’s really about handling multiple workloads in a single array, which is obviously very important in an Infrastructure as a Service model.  Virtual Domains do allow for different subgroups for the creation of storage by different organizations sharing a single array.The magic in a 3PAR array is in a custom ASIC that handles all the data movement within the array.  The controller processor handles all the control and data processing, allowing the ASIC to be built for a single purpose. Each ASIC is also direct connected (at 2Gbps) to the ASICs on the other controllers, allowing data movement to be striped across multiple controllers.  The ASIC is also built to handle the RAID5 XOR processing, virtually eliminating the write penalty when using RAID5.
  • Efficient
    3PAR arrays utilize a 16k block size, far smaller than most arrays, and a 1GB chunklet, allowing for a finer grained autotiering algorithm.  When thin provisioning, there is now preallocation, so the data written is truly the only data allocated to the LUN. Snapshots are handled with a redirect on write methodology, which they claim will allow 100s of snapshots on a single LUN.
  • Autonomic
    Adaptive Optimization (AO) is their autotiering engine.  HP has found that 90% of IOPS is in 10% of the capacity, which is why autotiering is such a hot topic in storage arrays lately.  The most common disk grouping 3PAR arrays contain are RAID5 SSD, RAID1 FC, and RAID6 near-line SAS.  I’ve heard from several 3PAR engineers that they recommend not initially installing SSD since most RAID1 FC disks can handle most customer workloads. Within AO, an administrator can define which tier a write is initially placed in.  Policies can then autotier the block up or down.Dynamic Optimization (DO) allows for the changing of an entire LUN or disk group to a new tier.  This is often used to migrate a LUN from RAID1 to RAID5.

Since I have customers who are debating between HP EVA and 3PAR arrays, I asked JR to comment on when one would make sense over the other.  He stated, with very little hesitation, that he wouldn’t buy an EVA at this point.  Granted, he did come to HP with the 3PAR acquisition, but I’ve never heard any HP employee admit the defeat of the EVA so directly, especially in such a public setting.  He backed his statement up with the fact that almost all HP Storage R&D is being funneled to the P10000, P4000 and StoreOnce products.

We were then able to work through a 3PAR lab.  While the 3PAR interface wasn’t convoluted or difficult to navigate, they could take some lessons in simplicity from the P4000 and EVA management interfaces.

Lab Tour

We were then taken on a tour of the HP Storage team’s lab environment, including seeing all the products we discussed earlier in the day and the power/cooling elements of the data center.  Some very cool stuff in there, which you can see in the following pictures.

File Storage

We finished up the presentations for the day with the X-series of products, the provide file-based storage on the IBRIX platform.

Jim Hankins, product manager for IBRIX, and Chris Duffy, product manager for file storage, kicked off the presentation with an overview of the product line.  This includes the X1000 (integrated), X3000 (gateway), and X5000 (integrated), all of which are Windows Storage Server based and include file and block storage, deduplication, snapshots, and file classification

Mark Tomas, IBRIX development manager, then covered the X9000, which is designed for archiving and massive scale out.  The top of the line is the X9720, which can contain up to 1.2PB using 2TB MDL-SAS disks.  A few of the features include:

  • Uses MDS shelves and blades (similar to P4800)
  • Single namespace across all models allowing tiering
  • A single X9000 namespace can grow up to 16PB
  • Tiering is based on metadata
  • Continuous remote replication is available to a node in the same cluster or another cluster

Robert Thompson, an Architect for Windows NAS Engineering, presented the X5000, a converged network storage server.  This is very cool technology and is the best example of converged infrastructure I’ve ever seen.  Imagine a 2U device that contains a couple of BL460s and a small drawer of disks containing 16LFF disks.  The blades run a Windows Storage Server cluster and direct attach to the disks in the drawer.  Up to 4 HP D2000 disk shelves can be direct attached to the device allowing for a maximum of 128TB with 116 disks (16LFF + 100SFF).  The possibilities for this hardware platform are very interesting (an E5000 Exchange appliance has already been released) and HP is currently listening very closely to customer feedback, so let them know your thoughts (think branch office vSphere cluster).

The second day started with a VMware integration presentation by fellow vExpert Eric Siebert and Aboubacar Diare (another guy I’d like to get to Omaha for some presentations).  Currently, the only gaps in VMware integration within the current HP Storage portfolio are VAAI in the P6000 (in QA right now), VASA on the P2000, and SRM 4.1 & 5.0 support for P2000.  If you want to use VASA with your HP storage, it requires the Insight Control Storage module or 3PAR Management Plug-in, which both act as the VASA provider.  They will then provide vCenter the array type, LUN type, snapshot information, whether the LUN is thin provisioned, CA relationship, and replication state.

They discussed how, due to Zero detect and Write Same features in the 3PAR, Eager Zero Thick (EZT) VMDKs on a 3PAR LUN can actually be faster to create (definitely not slower) than Lazy Zero Thick (LZT) VMDKs.  Of course, EZT VMDKs will provide better performance on writes to the VMDK once created.  For XCOPY & Write Same operations, the 3PAR, EVA, P9500 platforms all have abilities to throttle VAAI operations so as to not harm front end IO.


We finished up the presentations with Mike Koponen who presented the BladeSystem product line, which are designed for a quick and easy implementation of the hardware for a vSphere, XenServer or Hyper-V environment.  The line includes the following:

  • VS1 – Modular scalability, built on DL360s and P4500 and supporting VMware or Microsoft.  Both the servers & P4500 start at 2 nodes, and both can scale out to 8 nodes each.  The vCenter server and the P4000 Centralized Management Console are both virtualized on their own DL360.
  • VS2 – Density & flexibility, built on BL460s and P4800 and supporting VMware or Microsoft.  It starts with 6 blades, but can scale to 12.  The P4800 can contain one node (42TB) or two nodes (84TB).  It is also upgradable to CloudSystem.
  • CV2 – Similar architecture to the VS2, but offering both SAN & DAS versions, is designed specifically for client virtualization, and supports VMware or Citrix environments.
  • VS3 – Scale & performance, built on BL460s and 3PAR and only supports VMware.  It starts out with 16 blades and is scalable to 64.  The storage is built with a F-series 3PAR and starts at 38.4TB and can be grown to 162.2TB.  It is also upgradable to CloudSystem.

All four models include base and extended SKUs.

Overall, a great experience. I learned a bunch, and hopefully this post will help you out as well.

A big thank you to HP for putting this together and inviting me to join!

For additional coverage of the conference, check out Calvin Zito’s blog for his posts and links to the other attendee’s posts.