3PAR Mega News Bundle – Including Compression

3PAR Announcements

Today HPE announced a significant number of enhancements to the 3PAR product plus some changes in how the product is owned. The feature enhancements are enabled with the upgrade to 3PAR OS 3.3.1 which has been announced today. This really is quite some list, so hold onto your hat’s and let’s start with some data reduction enhancements which in combination HPE is calling Adaptive Data Reduction


3PAR has had dedupe for some time but has not had compression available until the release of 3PAR OS 3.3.1.  Compression is going be available on flash disks only and will be available to the GEN 5 systems i.e. the 8000 and 20,000. The aim of dedupe and now compression is to reduce the data size as much as possible on flash to make flash more affordable. Compression operations are performed inline i.e. before the data hits the SSD’s and are enabled at a per volume level. The best news is that compression is going to be licence free! HPE are expecting a 2:1 data reduction from using compression. I am going to do a deep dive on the compression and other data reduction technologies in the next few days, so watch out for that

Data Packing

3PAR writes to SSD’s in 16K pages, with compression of course you end up with odd sized pages.  These sub 16K sized pages are not optimal for writing to SSD and would incur a post process garbage collection to neaten things up and optimise their layout.  To eliminate this need for the additional garbage collection over head the odd sized pages are stitched together, just like your Nan makes a jumper, to form neat 16K pages. This process of taking odd sized compressed pages and packing them together is shown below.



Dedupe gets a code update in 3PAR OS 3.3.1. This new update changes the way dedupe operates writing to a private area first, as opposed to a shared area previously.  This change in operation aims to make the process more efficient, reducing garbage collection and ultimately making the system more scalable. TDVV (Thin Deduplicated Virtual Volumes) are depreciated, now you just create a standard volume and just change the attribute for the volume to turn on dedupe. Setting the dedupe attribute can be done from the CLI or SSMC.  Dedupe and compression can be combined together and HPE are expecting a median 4:1 data reduction, when the technologies are combined. The good news is that dedupe is again licence free and can be applied to all the GEN4 (7000 + 10,000) and GEN5 (8000 + 20,000) systems.


I really like the idea of VVOLs and have covered them in depth previously.  Their take up has been a little slower than I would have expected and one of the main reasons I have seen cited was the lack of replication support.  vSphere 6.5 introduced VVOL replication and 3PAR now supports this functionality once upgraded to 3PAR OS 3.3.1.  Don’t forget given the granular nature of VVOLs this will mean that replication can be controlled at a per VM level.

Peer Persistence 3 Data Centres (3DC)

Peer Persistence has to be one of my favourite 3PAR features, this enables the creation of a metro cluster or stretched cluster if you like across data centres.  This high availability setup allows storage to be switched between data centres, online with no disruption to hosts.  Peer Persistence 3DC allows a 3rd leg to be added to the replication topology.  This 3rd location does not become part of the metro cluster but allows a 3rd copy of the data to be replicated asynchronously for data availability.


Free Licences

3PAR must have more native software allowing data services than any other SAN now.  Deciding which ones you wanted and then having to pay for them was painful. Well now all licences within a single system are free! If you have more than one system and are using features that allow the systems to talk to each other e.g. Remote Copy, Peer Persistence, Peer Motion you will need to upgrade to the multi-site licence.  Again once you have the site licence this enables all multi system functionality.

Replication to StoreVirtual VSA (Peer Copy)

I have been wondering for a while if and when they were going to enable replication between the StoreVirtual and 3PAR, well it is now possible with the use of RMC in a feature called Peer Copy.  RMC is now included in the free licence bundle and in this circumstance acts as the data mover, replicating crash consistent snapshots between a 3PAR and Store Virtual.  Other RMC enhancements include replication to Azure, deployment in Hyper-V and SAP HANA support.


Online Import

The online import tool which already supported the EMC Clarrion, VNX and VMAX now adds support for the EMX DMX. Other vendors already supported include Hitachi and IBM

File Persona

File persona gets a bunch of feature additions including doubling scalability, file locks for governance, cross protocol file sharing and a file system checker.

Extended SSD Warranty

The existing warranty on 3PAR SSD’s is 5 years, now this has been extended to 7 years.  This covers media wear out and electronics.  The offer is open to all SSD’s in 8000 and 20,000 systems bought after June 2015.

There are even more enhancements than this but I don’t want this blog turning into a hardback edition so I will bring more news in the coming days on other enhancements and deep dives on what has been discussed today. To make sure that you don’t miss any updates, you can get e-mail updates, or follow via Face Book, LinkedIN and Twitter.

This is a mind blowing amount of information and if your head is spinning check out these nice videos in which Calvin Zito summarises these announcements in two videos Video 1, Video 2.




3Par Gets Flashier

For a while standard 3Par owner’s could be forgiven for looking on with envy to those lucky enough to own the all flash 7450 model, with its all flash massive IOPs crunching potential and new features such as dedupe. Well feel jealous no more 3Par fans, last week HP shared out the flashy goodness to the hybrid arrays by announcing a bunch of features that will well and truly pimp out your 3Par, to an extent which would even impress Xzibit. Check out this orange Roller below, I bet Rolls Royce didn’t see that coming when it rolled off the production line.



3 Par OS 3.2.1

The announcements from HP centred around using flash to deliver an extra whack of performance for your 3Par. The key new features were in summary Adaptive Flash Cache, Dedupe and Express Writes. All these new features are enabled by upgrading to 3Par OS 3.2.1, and the even better news is that all these new features are freebies i.e. they are included as part of the standard 3Par OS. Let’s break each one of these elements down and see how it is going to deliver additional performance.

Adaptive Flash Cache

Adaptive flash cache is about utilising SSD’s to expand the size of the cache. A bigger cache allows more data to be stored in cache and hence a greater likelihood of being able to retrive data from cache. This is great news for anyone’s environment that has high random reads, judging by the fact my most popular post of all time is still Adaptive Flash Cache Deep Dive lots of other people agree. Do check out the deep dive it’s got lots of good info in it. Also check out this Adaptive Flash Cache video Calvin Zito has put together where he shows a practical demo of the kind of performance improvement that is possible.

Express Writes

Next up Express Writes, again like Adaptive Flash Cache this new feature is aimed at improving latency but for writes by optimising the FC protocol. Express writes aims to deliver lower latency by delivering lower CPU interrupts per IO, Performance improvements can be up to 10%.


Dedupe was announced earlier this year for the 7450, and last week’s announcements was that it will now be available to all 3Par systems with the Gen 4 asic i.e. the 7000 and 10,000. The way the dedupe works is exactly the same as on the 7450, inline and by the assigning a hash to each unique incoming write and then comparing the signature of further incoming writes to ensure they are unique. The dedupe process is demonstrated in the diagram below taken from the HP whitepaper HP 3PAR StoreServ Storage: optimized for flash. The limitations to using dedupe are that it is only available on the SSD tier and the technology cannot be combined with AO.

Final Thoughts

This new set of features has to be one of the most compelling reasons to upgrade in some time. I’ll be upgrading ASAP and really making those SSD’s work for their living!

You can see a summary of the announcements in this ChalkTalk.


Rumors Update

Previously on 3pardude.com I posted a link ‎through to Chris Mellors’s post HP Discover 3Par round up. The two key rumors were dedupe would become available across the 3Par range and that flash cache would be introduced.


Last week HP’s Calvin Zito, their online marketing guru and purveyor of YouTube videos with catchy tunes made a post on the HP StoreServ forum, his post went as far as confirming the rumours in an unofficial way and that’s official 🙂


Here are the important bits from the post:


  • We don’t have Cache via SSD yet – but stay tuned
  • At HP Discover, we also discussed that we’re using the same ASIC in the 7450 that is in the 7400, 7200, 10800, and 10400.  I’ll let you read the tea-leaves.


You can read the full post here.


Dedupe in the 7450 was previously discussed in an earlier post . It’s no surprise to see this is to be introduced across the range, as a diversion from this would have taken away one of 3Pars key selling points that all features and management are common to all models. Dedupe combined with class leading thin provisioning could lead to a very small foot print indeed.


With flash cache as it’s still not officially confirmed details are still sketchy, but essentially it is likely to be a technology that leverages SSD’s in the system to add a second layer of caching to the existing controller cache. The advantage of this will be a larger target for active data to ensure that this hot data is being served from controller cache or SSD’s. This to me would plug the gap of unpredictable work load patterns that AO misses. AO works by analysing a sample period and then moving data between tiers based on these patterns once the analysis period finished. Whilst this works well for predictable workloads that follow a pattern over time any workloads that do not follow a historic pattern would leave AO too late to the party and moving data that is no longer hot.


This will make for a really interesting fight with the other mid-range arrays from EMC and Netapp which already offer their own versions of dedupe and SSD caching.