Monday 28 January 2008

Controls in Place

I work with many companies in which the storage and server/platform teams have to sign in blood in order to receive suitable access to perform their day to day work. In fact, for some organisations, it can equate to 50% or more of their time. I've been there and I understand and have experienced the frustrations it creates.

Compare this to our trader friends. Jerome Kerviel allegedly managed to gamble €50bn, more than the assets of his bank, Societe Generale. I wonder how many change records he had to raise in order to get access to that level of funds?

Sunday 27 January 2008

HMRC Get Better And Better

Following on from the "inadvertent" loss of 25 million personal records, it appears that good old Her Majesty's Revenue and Customs have decided that online tax filing is not as secure as they may have indicated in the past (see BBC link). High profile celebrities, MPs and the Royal Family will not be allowed to post online. I have a feeling that HMRC will be moving to force all of us in the UK who submit a tax return to submit the return online in the future, so the implication is that it's OK to lose the records and compromise the personal and finanical identity of the general public (without recourse, by the way).

Is data security that bad? I really hope not, as our future depends on it.

Monday 14 January 2008

My Drobo Has Been Ordered

I've been mulling over for some time whether to purchase/evaluate the Drobo storage device. Up to now I've been put off by the lack of NAS support, as the device only supports USB2.0. However all that has changed; I read today that there's now the DroboShare, which converts the Drobo into a NAS device. I've yet to see the full details (and I doubt the DroboShare will not be available some time in the UK) however this now justifies my purchase (as no-one has offered me one so far!).

The order is in - for Drobo and two 1TB drives. I can't wait...

EMC Stretches DMX Both Ways

EMC made two DMX-4 announcements today. At the top end, the first details solid state drives which replace standard spinning HDDs. At the bottom end the second announces new 1TB SATA-II drives (and other miscellaneous stuff).

The solid state hard drive (or as Chuck Hollis describes it, the enterprise flash drive), isn't just a small 1-2GB device; this is a real replacement for 73GB or 146GB drives and will apparently slot right into a standard HDD bay. Assuming drive support is identical to standard HDDs, then it should be possible to create hypers from the drive(s) and present them to many separate hosts. I see this as significantly better than products such as Tera-RamSan. Firstly, the EMC device is more compact, second and most important, it integrates with the existing array, meaning better compatibility, support for existing functionality such as SRDF and TimeFinder and therefore more consistency.

Obviously a real issue will be cost versus standard HDDs, however I think there will be plenty of takers. Every large organisation has one or more application which needs more storage horsepower and the tradeoff between rewriting the application, purchasing more hardware or using SSD HDDs in DMX will make the latter option very appealing. Hopefully as time goes on, the cost of SSD drives will reduce significantly (think of how expensive SD cards used to be) and SSD technology will be more attractive to a wider audience.

Naturally the blog spin masters at EMC have been active; here's just a selection of the links I found:

Nothing yet from The Storage Anarchist though. Give it time...

The second announcement covered availability of 1TB for DMX-4. Fair play to EMC, they are offering storage technology across the spectrum of speeds and capacities, all in a single box. I still need to think through the implications of having even more choice in one array, not least of which is the impact on my design methodologies!

Oh and the second announcement re-announced thin provisioning for DMX-4. But everyone else is already doing that already. :-)

Saturday 12 January 2008

I Stand Corrected

In a previous post earlier this year I mentioned the Onaro purchase by Network Appliance. As I said at the time, I wasn't aware Onaro's SANScreen product even had a NAS module. It seems I was wrong, and thanks for Deni O'Connor for indirectly pointing it out. In fact, SANScreen now has NAS Insight which provides for NAS monitoring support. However this feature was only made general availability on 31 December 2007, so you can hopefully excuse my oversight for not realising it has been released.

(On a side note, why are large Enterprises such as Onaro still not using RSS to announce product releases? I haven't got the time or inclination to trawl their websites each day. RSS is so much easier.)

I had a quick look at the NAS Insight press release and details on their website. Although there's a demo, I couldn't ascertain what NAS products (other than Netapp filers which are in the demonstration) the product supports. That makes me think it supports nothing BUT Netapp (although I again stand to be corrected). If that's true then NAS Insight is a pointless feature for many customers who would want to use the product for cross vendor consolidation and certainly doesn't demonstrate Onaro's NAS credentials. Compared to the last release of DFM I saw, NAS Insight is pretty poor.

To date my SANScreen exposure has been based on one large "global installation" of the product and presentations from the Onaro marketing team. When an instance of SANScreen was enabled in one location of the global deployment, it created thousands of exceptions which then required manual intervention. When I last had a presentation on the product (in October) there was a large number of SAN scenarios SANScreen wasn't reporting on, a lot of these relating to non-EMC and replication support. There's still a long way to go yet.

Up to this point, Onaro may have developed relationships with the vendors which provides for ongoing access to new releases of their management tools in order to extract configuration information. Going forward, will those companies still be as keen to provide that information to Netapp, who may be their direct competitor in the NAS (and non-NAS) marketplace?

Thursday 10 January 2008

Memory Lane

I thought I would spend some time tonight looking back at my blogging history. I first started writing stuff on a website I created called Storage Eye. The first "article" I wrote was posted in February 2002, almost 6 years ago. It was pretty crude stuff (the presentation, not the content), simply putting static pages up on a website. I followed that up with postings in 2004 on ITToolbox and then in 2005 back to my own website (this time database driven), eventually cutting over to blogspot at the beginning of 2006.

My material hasn't changed much over time. The main difference is probably that I post more often and with longer content but that it probably because the tools available now make that really easy.

What has definitely changed are the tracking tools and the ability to locate bloggers who are out there. I track 69 storage blogs using RSS (and quite a few more non-storage related). Tools such as Technorati, Digg, Stumbleupon, allow blogs and their feeds to be tracked and rated. The blogosphere is active and vibrant and full of lots of great content.

Where's it all headed? Who knows, but I am sure it will be fun!

Monday 7 January 2008

XiV Part II

Following on from BarryW's comment to my XiV post, I've been thinking over how the XiVarchitecture works. When a disk fails and the missing mirrors need to be recreated, then the data is likely to exist across all or most of the configured drives. Logically it would make sense that the target for the new lost mirrors would be all drives. So, as a drive fails, all drives could be copying to all drives in an attempt to ensure the recreated lost mirrors are well distributed across the subsystem. If this is true, all drives would become busy for read/writes for the rebuild time, rather than rebuild overhead being isolated to just one RAID group. Whilst that seems like a good thing for rebuild time, it seems like a bad thing for performance. Perhaps this isn't the case and in fact the failed device is re-created on a spare drive by copying all the mirrors back in from their other location.

Following the same line, in order to recreate a failed drive and rebuild the lost data across the array, then each drive must have spare capacity; in say a 200 drive system, that would require about 1/200th of all drives to be free at any one time, ready to receive rebuilt mirrors. Obviously the alternative option is just having spare drives, but that sounds less interesting!

What about when the failed disk is replaced? There must be another algorithm which ensures the replaced disk is not a target for all new writes, so presumably, static mirrors are pro-actively moved onto the replaced device.

This architecture throws up some interesting questions, especially when trying to understand performance. I am starting to get excited about messing about with one!

2.5" is cool

Hitachi Global Storage Technologies (the old IBM Disk Division) have announced a 500GB 2.5" drive for laptops. Not particularly interesting I hear you say, however these latest drives from HGST now come in Standard and "Enhanced Availability" versions. The enhanced version claims to be able to run a 24/7 workload rather than standard day to day access required in a laptop.

So does this mean these SATA drives are suitable to use to build an array? Well, first of all they are not fast ast 5400RPM. It would also be nice to have the SAS rather than SATA interface, but just look at the power consumption figures; 1.9W in read/write mode!

How about IBM taking the XiV technology and using 2.5" drives with this power consumption and capacity - now that would give EMC something to complain about!

Thursday 3 January 2008

Two for the price of one

The holidays are over and it's back to work for me. In fact I returned yesterday; the break was good however it is also good to be back.

It seems that I've returned to a flurry of acquisitions. Yesterday there was the heavily reported (on the blogosphere) purchase of XiV by IBM. Tony Pearson gives a summary of the features on his post. One thing that interests me is the use of distributed writes across an entire array by creating 1MB blocks from (presumably) LUNs and filesystems. If a drive fails, then the data is still available on other disks in the system and spread across a great number rather than a single drive (RAID-1) or potentially a small number of drives (RAID5/6).

I've been trying to get my head around what this means. On the one hand it sounds like a real problem, as a double drive failure could impact a wide number of hosts; it all depends on how well the 1MB chunks are distributed. However maybe it isn't that much of a problem as the issue only arises when both of the chunks that mirror a 1MB block both occur on failing drives. I would expect that as the number of physical drives increases then the impact of double failure reduces, as does the number of 1MB blocks affected. In addition, a drive may fail only in one area rather than on the whole device, so the affected blocks could be quite small; the remainder could be perfectly readable and be quickly moved. No doubt Moshe and the team have done the maths to know what the risk is and compared it to that of standard arrays and wouldn't be selling the product if it was not inherently more safe.

The only other issue I can see is what market the product will slot into; Tony mentions that the product is not for structured data (although I guess it supports it) but was designed for unstructured data of large binary file types. So, why use RAID-1 compared to say a 14+2 RAID-6 configuration which would be much cheaper in terms of the disk cost? Presumably another selling point is performance, but I would expect the target data profile (medical, large binary objects) to be more sequential than random access and not be that impacted by using SATA.

I guess only time will tell. I look forward to seeing how things go.

The other purchase announced today was that of Onaro by Netapp. Onaro sell SANScreen, a tool to collect and analyse fibre channel SANs and to highlight configuration issues. Whilst I think it is a good product, I don't see the fit with Netapp's business in the NAS market (in fact I'm sure SANScreen doesn't currently support NAS), so where's the benefit here other than buying up a company which must be close to or is making money.

I wonder who will be bought tomorrow?