the storage anarchist

2.043: storage wrappin' about tiered storage

This just in…a particularly timely episode of Storage Wrappin' (source unknown)

(If you cannot see the embedded video, click here)
 

technorati tags: 3PAR,auto-mneumonic tiered storage,Hey Now!,Marc Farley

2.042: bring out your dead!

My, what a week already.

IBM finally got around to putting the still-borne DS6800 out of its misery – something I had thought they were smart enough to do over two years ago (I was apparently wrong). Not to worry, I guess – if you really want to have one of these useless beasts, I understand they are still available over on e-bay.

Once touted as the entry level Shark, the DS6800 was purported to share the vast majority of its code with the higher-end DS8000 series. Over time, it became clear that no such miracle had been performed – the DS6800 was even less feature rich than the DS8K. And with the brandy-spanking-new DS8700 lacking several features that were touted as foundational for the DS8000 platform family (e.g. thin provisioning, LPARs and the like), it has got to make you wonder how serious IBM is about this space.

But undoubtedly attracting the most attention has been the comments from NetApp's CEO Tom Georgens late last week that the notion of storage tiering is dead.

There has been a lot of Twitter chatter about Tom's assertion, and at least a few blog posts - e.g., Mark Twomey's (@StorageZilla) Virtual Vs. Static Provisioning. Martin Glassborow's (@storagebod) The Crying Game, and Chris Evans' (@chrismevens) Enterprise Computing – Death of Tiering?. And even today the debate rambles on in Twitterville, with Alex McDonald (@alextangent) in the middle of the debate over whether PAM II + SATA is "tiering" or simply "caching."

All good fun, but I'd like to bring forth a slightly different perspective for why there is more to tiering than simply Flash and SATA.
 

2.041: pack light!

OK, I don't usually do this, but what the heck…someone sent me this today, and I couldn't resist:

Coming from me, this is inarguably FUD, irrespective of whether there is any relationship to the actual truth.

But you gotta admit it is kinda funny .

technorati tags: NetApp,NTAP,High Overhead,Low Performance,Zero Scalability

2.040: dmx gets some fast love!

(Title intentionally modeled after Chris Mellor's articles on The Register)

Last quarter EMC gave V-Max customers a couple of pretty special Christmas/Holiday gifts. First, there was FAST for Symmetrix V-Max, followed closely by new hardware (8Gb FC/FICON) and several new software features.

This week, in a timely show of love for DMX customers (Sunday is Valentine's day, after all), EMC delivers a software update for the DMX-3 and DMX-4 platform that includes FAST (Fully Automated Storage Tiering) for Symmetrix DMX.

Like FAST for Symmetrix V-Max, FAST for the DMX platform automates the movement/relocation of Open Systems LUNs and CKD Volumes across different storage types based on performance requirements. FAST can both promote data to faster storage types (like flash drives) to maximize the ROI, and it can demote data to other storage types (like SATA drives) to leverage lower-cost/high-capacity storage for less-frequently accessed data. With FAST, systems will typically require fewer total drives to meet their performance and capacity requirements, making for a smaller footprint and more energy-efficient storage solution.

 

2.039: tid-bits and micro-bytes

I thought I would try something new, just for a change of pace. Rather than my usual heavyweight blogs, this post is a collection of "sound bites" that have been floating around in my head. None really large or meaty enough to justify my usual deep dive, but things I just can't let pass by without comment.

scott brown (r) ma

I'll start with this one, if I may:

Yes, I am a resident of Massachusetts.

Yes, I'm a registered Independent.

And yes, I voted to hit the Reset Button on Washington, DC.

You're welcome, America.

(And no, I didn't make any money from the Google Ads that Scott Brown placed on my blog during his campaign.)

storage guarantees

Almost everyone can reduce their storage requirements, and almost every vendor has product features that will help (here's my take on how to make it happen). The amount you will save will always depend upon how you've been managing and deploying both storage and applications in your environment. But there is no simple answer, thus all the caveats, prerequisites and exclusions in the recent rash of guarantees.

(hey, I'm just getting started – read on)…
 

2.038: the anarchist's universal storage guarantee*

 

I, the storage anarchist, do hereby guarantee to any and all information storage consumers, irrespective of such consumer’s chosen storage vendor or supplier, and/or said consumer’s geographic locale, and, without limits as to time or space, the following:

1. Your storage requirements for a given set of application(s) currently deployed on RAID 1 will be reduced by approximately 43% if you convert to using RAID 5 (7+1) or RAID 6 (14+2). If you choose a different source or target RAID protection ratio, your savings will vary. Your savings percentage can be calculated as 1-(NEW/OLD)%.
2. If you currently use “fat” provisioning and over-allocate capacity by 30% for a set of applications, you will reduce your storage requirements by 30% if you convert that capacity to a “thin” provisioned logical storage device(s). If your storage admins have been over-allocating more than 30%, your savings will be larger; if they have been carefully optimizing their provisioning and over allocating less than 30%, your savings will be less.
3. Your storage requirements for file-system based data can be reduced to only that capacity currently in use by objects stored in that file system if you convert the logical storage device containing that file system to a thinly provisioned device – but only if you are not already using thin provisioning.
4. “Zero space reclamation” features of thin provisioning can recover unused space and space that is specifically released by the host operating system/file system; however, on some popular host/server platforms (e.g. Windows), space for deleted files is not typically released or zeroed, reducing your total opportunity for savings.
5. Consolidating the storage from multiple applications, servers or “virtual machines” can reduce your capacity, CapEx and OpEx requirements vs. direct attached storage installed in your servers/hosts, especially if those applications have capacity and/or retention requirements that are not optimal for so-called “captive storage.”
6. If 10% of your data for a given set of applications changes between backups, you will require 90% less capacity for those applications’ point-in-time backup images if you use logical snapshots instead of full-volume clones. If more changes, your savings will be less.
7. Your on-line and/or backup storage requirements will be reduced by up to 50% through use of data compression, dependent upon data type, content, prior compression, etc.
8. If you routinely store and/or backup a large number of similar data containers (e.g., system/boot images, shared documents, development clones of production data, etc.), your physical storage / backup requirements will be reduced to 1/Nth of the total logical capacity of those containers (where “N” is the number of identical logical copies of the objects in those containers).
9. Archiving of information no longer being accessed but for which extended retention is required (e.g., by government or industry mandate) can significantly reduce both on-line and backup capacity requirements.
10. Your storage costs (CapEx and OpEx) will be reduced if you store your infrequently accessed data on larger capacity drives (e.g. large SATA drives), your frequently-accessed data on solid state storage (e.g., cache, DRAM, flash, etc.) – done right, you will be able to eliminate the most expensive storage device from your infrastructure in combined terms of $/GB and $/IOPS: the 10K/15K rpm enterprise disk drive.
11. If you use larger and/or faster drives as described, you will also reduce your total power, cooling and space requirements for storage.
12. Additional types of persistent storage may further reduce your costs, including both on-premises solid-state storage devices and the emerging availability of large-scale low-cost capacity made available by so-called “cloud” service providers.
13. You will further reduce your OpEx if your storage platform eliminates the need for time-consuming human intervention by autonomically optimizing the utilization of applied DRAM, Flash and SATA to meet your defined service level policies and objectives.
14. If combine multiple of the above conditions to a specific set of application(s), your savings will increase, although the savings may not be additive nor multiplicative.
15. Variation of actual savings will likely vary between vendors/suppliers, but any difference will generally be insignificant or temporary. The dynamics of supply and demand will continually drive vendors/suppliers to deliver similar solutions; thus market differentiation is most likely not to come from specific space-saving features themselves in the long run.
16. Oh, and yes: your information storage requirements will continue to increase year over year, ad infinitum (or as long as governance dictates, whichever comes first).
17. Your performance, availability, business continuity, disaster recovery and/or data retention requirements may limit your ability to realize one or more of the above savings potentials <thanks, Alex, for pointing that one out>

I personally guarantee it.

* excess installed capacity may be required to support future growth, and is therefore excluded from this guarantee.

 

Feel free to offer improvements, corrections and additions – I will update this Guarantee so long as the assertions are generally universal.

Change Log
2010-02-04 10:04AM EST: Added #17 in response to Alex's suggestion (below).
2010-02-04 04:00PM EST: Corrected syntax errors in #9

technorati tags: EMC, storage, guarantee