Archive for the ‘Gestalt IT’ Category

EMC’s Unified Platform and Storage Tiering on GESTALTIT.COM

September 24th, 2009 No comments

Check out the new featured blog post on GestaltIT by myself and Ed Saipetch on EMC’s Unified Platform and Storage Tiering, where we talk about some important next generation features and products from EMC. Also discussion revolves around possibly the direction EMC is moving with their Unified Hardware – Software Platforms and Storage Tiering using next generation technologies like FAST.


Cloud: The Quest for Standards

September 17th, 2009 No comments

Note: This is my first attempt to write about the Cloud technology, please feel free to correct me if my understanding of any of the aspects is not correct.

Many of us in the industry think about Standards as a hindrance to technological growth. Creating Standards has been a long drawn process for any technology that is set for robust growth. Cloud Services is one of those areas where we expect an exponential growth over the next few years. It’s better to control the behavior of growth rather than taming or enabling haphazard growth.  Standards typically help and are known to work in the favor of the consumers.


Why Standards


Everyone in the industry or home users heard about the battle between HD-DVD and Blu-Ray, years of consist & persistent fight between two major groups. Toshiba and alliances supporting the HD-DVD format while Sony and alliances supporting the Blu-Ray disc formats.

Billions of dollars of investment, millions in marketing, millions in legal battle, millions in customer investments, millions in HD-DVD investments, futures & employment dependent on those technologies.

In the end, HD-DVD loses the battle, Blu-Ray wins. Well who else lost, the consumers, the investments, the video game developers, the alliance partners and that means all of us. All the way around, YOU as a consumer is never a winner in this battle.


Lets look at the WIFI technology 802.11 a/b/g/n. When all these WIFI technologies started, how excited were we, wow now I can connect my laptop over the air and browse the Internet and do not need a physical cable.

Imagine not having the 802.11 standards in place during this boom. You buy a Linksys WIFI for your home and install a Linksys Air card in your laptop and all works good. You go to the Starbucks around the corner and they have Netgear and your laptop WIFI stops working. You go to the office and they have a complete separate set of devices that are incompatible with your existing WIFI. All around it would have been chaos.

Wow, its good we didn’t go through that. Why, because there were standards. Interoperability was the key. Everyone made products, brought enhancements, but at the end of the day, they were all compatible with the users or consumers or customers.



A Standard is typically a document that defines a certain set of shared protocols, resources, API’s, interoperability, security, methods, practices and other aspects relating to usage.

A Standards committee is normally formed with various manufacturers, service providers and experts that act as governing bodies defining and deciding on common technology practices.

Though do not get me wrong, Standards will not define your product features.

The Cloud Standards

Though some experts differ on the opinion for having a Cloud Standard early enough in the game, which is right now. I truly believe, in the best interest of the consumers, customers and investments, a common governing body should start deciding on the standards for Cloud Services.


Hypervisor is the underlying technology, which enables virtualization in the cloud.

Question: Can a Standard be incorporated to enable applications to move from Hypervisor to Hypervisor (Vmware vSphere to Citrix Xen or Microsoft Hyper-V) without going through a redesign?

Result: That will enable your applications to move around the Cloud infrastructure irrelevant of the underlying technology provider.

Today: This cannot be done

Service Provider

Some of the famous and emerging Service providers today are Amazon, EMC Atmos (Beta), Nirvanix, Terremark, Savvis, Rackspace.

Question: Can you move your applications from Amazon to Rackspace and how easily can this be achieved.

Result: Free movement of the application irrelevant of the underlying Service provider, very hard to achieve.

Today: This cannot be done


Security within the Cloud Infrastructure enables user, applications, application owners and other related applications to interact with each other based on standard protocols of communications.

Today: There are no standard security protocols used within the Cloud Infrastructure that are common amongst all service providers. Every security centric provider and service provider alliance will enable you with a certain security mechanism.


Underlying storage may ways be different, no real dependencies on use of Storage as it relates to Cloud and Standardization. Though Amazon uses a certain set of Storage, while Terremark and Savvis may use a certain kind, it should not create dependencies on application movements.


Application movement between private clouds. Application movement between public clouds. Application movement between private and public clouds.

Question: Can you move your applications from Terremark to Amazon’s EC2 and then back to the private cloud.

Result: The re-engineering of the application and the related API’s will enable that movement

Today: Very hard to do, or a very tedious process, No standards around it today.

Cost Models

Yet another driving force with Cloud based services. Some Service Providers today charge based on computational power, some on storage capacity, some on bandwidth, some on a mix of all

Question: Is your application hosted in Amazon’s EC2 costing you the same money that a similar application at Terremark would cost you.

Result: Standard pricing practices needs to be established. Pricing can vary based on feature sets.


Standards vs Dominance

So if you enable a certain technology to grow by itself in an uncontrolled manner, it just becomes a dominating factor and tends to mold people, industry, customers, ideas all in one direction. Typically that direction is where the dominance is.

If we wait too long for an establishment of a Standard for a certain technology, more or less that technology will be the Standard and everyone will need to adapt around it. Have you heard “Survival of the fittest and Dominance of the Strongest”?

The Customer

Remember you are talking about large scale investments with Cloud Services being developed, designed and implemented.

We should see exponential growth with the Cloud Services market over the next few years. This means billions of dollars will be poured into R&D, Infrastructure, Services and other related areas enabling Cloud Services.

This is the ripe time when things are yet not out of control, the more players join the mix, and the more dominance effect will persist.

This is the time to act and act in our favor to preserve the investment in technology that we all expect and hope to see growing. SNIA and DMTF is currently working on the Cloud Standards initiative (EDIT 09/17/09 at 11:15 AM).

It seems like as this post is being released, the first CDMI (Cloud Data Management Interface) draft is ready and released by SNIA. Glad to see the progress.

Unrelated to this discussion by a really funny Youtube Video by Mr. Larry Ellison of Oracle on Cloud Computing

Other Reading on Cloud (EDIT 09/17/09 at 11:15 AM)

Note: This is my first attempt to write about the Cloud technology, please feel free to correct me if my understanding of any of the above aspects is not correct.

SUN ORACLE Exadata Version 2: Showing the power of ORACLE SUN

September 16th, 2009 No comments

The US Regulators gave the ORACLE purchase of SUN a go ahead several weeks ago, but EU Regulators are still actively looking at Antitrust laws with the possible buyout of SUN by ORACLE. ORACLE’s (Mr. Larry Ellison’s) quest to own an infrastructure company is becoming true with the purchase of SUN.

But the approvals haven’t stopped Mr. Ellison’s Team from redesigning the SUN ORACLE Exadata platform Version 2 (With SUN ORACLE logo’s on it). A joint venture between ORACLE and SUN has been on for several years now. Today was the day when the new Exadata platform version 2 was presented to the world by Mr. Ellison himself. It was truly visible, that Mr. Ellison is already taking a lot of pride with this acquisition even before its approved.

There was an advertisement earlier this week from ORACLE SUN challenging IBM and all its products and how Mr. Ellison now wants to go after IBM to become the top Infrastructure company. Said that, there are only 3 big infrastructure companies today, IBM and HP going neck to neck in terms of revenues competing for the 1st position, while the pending approval of ORACLE – SUN at number three.

Through it is great to see the vision of Mr. Ellison and how he is internally transforming ORACLE from being a software database company to an Infrastructure company. Today’s announcement of SUN ORACLE Exadata version 2 platform is very unique in that sense. Exadata products has been developed with years of partnership between SUN and ORACLE, but goes to show how both the combined companies can fulfill the datacenter vision END to END.

This platform extensively uses the SUN FlashFire technology and is truly the first OLTP (Online Transaction Processing) System designed to optimize customer data processing using a mix of SUN hardware and ORACLE software. It was very noticeable during the 35 minutes introduction where Mr. Ellison drove the presentation for more than 25 minutes and then handed over to John Fowler, EVP SUN for a technical talk.

Clearly ORACLE is targeting IBM and NCR – Teradata products with the release of this platform. It was obvious in the presentation that Mr. Larry Ellison used the word “THEY” numerous times signaling towards IBM and NCR. Though it was not said during the presentation, “THEY” could include HP as well. At this point without the final approval of the SUN purchase, it wouldn’t make a lot of sense for ORACLE to make another enemy, HP.

Here are some SUN ORACLE Exadata Version 2 platform highlights

  1. Exadata Version 2 is optimized for OLTP (Online Transaction Processing), first in its kind to hit the market.
  2. Typically SUN ORACLE Exadata Version 2 should give customers a 50X to 100X better performance than standard data warehousing servers.
  3. Optimized for Random I/O
  4. 1M IOPS per system cabinet
  5. Each system cabinet has 8 Compute servers, 176 total processors, 336 TB of Raw Disk, 5TB of Flash Cache (56 Flash Cache cards), 400GB to total DRAM in the 8 compute servers.
  6. Intel Nehalem processors, Infiniband switching, FlashCache, 4 Ethernet links per database node.
  7. Runs Linux System, Oracle manages cache, fully redundant compute servers and storage. On demand capacity expansion as it relates to compute servers, storage or Infiniband switches.
  8. 1 Node (computer server) is the smallest configuration, large configurations can be 8 nodes in one cabinet or 32 cabinets combined together to massively have 32 Million IOPS or several 100 Petabyte of storage optimized for OLTP.
  9. Infiniband speed per link is 40 Gbps aggregating to 880 Gbps for a system (cabinet), Non Blocking switch gives a full open & distributed system access for faster processing.
  10. Power consumption less than Exadata version 1.0 by 14%
  11. Fastest OLTP system, Fastest Data Warehousing system in the world
  12. All calculations done in memory (FlashCache), optimizing the system.
  13. Massively Parallel Processing, the scale out architecture helps and enables easy on demand expansion.

Couple things to note:

  1. Mr. Ellison calls Flash Disk as Dumb Flask Disk, truly remarkable.
  2. Also another highlight was to use 56 Flash Cache cards per system (5 TB) and then use 168 x 2TB SATA drives (Possibly 7.2K RPM) to optimize data space (Somehow didn’t make a lot of sense).

Some Questions to Consider

  1. Is this a real threat to Storage & Host providers where you have specialized Hardware / Software combination optimizing your performance for certain applications?
  2. Is this the power of ORACLE SUN that we will see in the future?
  3. How does this compete with EMC COMPUTE platform (rumors) or the Cisco – EMC Alpine Project (rumors)?
  4. Does the VCE (VMware – Cisco – EMC) partnership really focus towards the giant to come ORACLE SUN?
  5. What will happen to the ORACLE – HP partnership if the ORACLE SUN buyout gets approval, what happens to HP – Oracle Exadata?
  6. Does this create any Antitrust scenarios for the future?
  7. Is Mr. Ellison’s dream to own the infrastructure end with the purchase of SUN?

Here are some links for references if you would like to read more about Exadata products

Symmetrix V-Max Systems: SRDF Enhancements and Performance

September 10th, 2009 No comments

v-max image 2So this was one of those posts that I always wanted to write related to Symmetrix V-Max and SRDF enhancements that were incorporated with the 5874 microcode.

Yesterday morning had a chat with a friend and ended up talking about SRDF and then later in the day had another interesting conference call on SRDF with a potential customer. So I really thought, today was the day I should go ahead and finish this post.

Back in April 2009 when the V-Max systems were initially launched, Storagezilla had a post on V-Max and SRDF features, he covers quite a bit of ground related to the Groups and the SRDF/EDP (Extended Distance Protection).

Here are the highlights of SRDF for V-Max Systems

SRDF Groups:

  1. 250 SRDF Groups with Symmetrix V-Max (5874) Systems. In the prior generation Symmetrix DMX-4 (5773), it had support for 128 groups. Logically even with 2PB of storage, very seldom do customers hit that mark of 250 groups.
  2. 64 SRDF groups per FC / GigE channel. In the previous generation Symmetrix DMX-4 (5773), there was support for 32 groups per channel.

SRDF Consistency support with 2 mirrors:

  1. Each leg is placed in a separate consistency group so it can be changed separately without affecting the other.

Active SRDF Sessions and addition/removal of devices:

  1. Now customers can add or remove devices from a group without invaliding the entire group, upon the device becoming fully synced it should be added to the consistency group (with previous generation Symmetrix DMX-4, one device add or remove would cause the entire group to invalidate requiring the customers to run full establish again).

SRDF Invalid Tracks:

  1. The “long tail” – last few tracks search has been vastly improved. The search procedure and methods for the “long tail’ has been completely redesigned. It is a known fact with SRDF, that the last invalid tracks take a lot of time to sync as its going through the cache search.
  2. The SRDF establish operations speed is at least improved by 10X; see the numbers below in the performance data.

Timefinder/Clone & SRDF restores:

  1. Customers can now restore Clones to R2 and R2’s to R1’s simultaneously, initially with the DMX-4’s this was a 3-step process.

SRDF /EDP (Extended Distance Protection):

  1. 3-way SRDF for long distance with secondary site as a pass through site using Cascaded SRDF.
  2. For Primary to Secondary sites customers can use SRDF/S, for Secondary to Tertiary sites customer can use SRDF/A
  3. Diskless R21 pass-through device, where the data does not get stored on the drives or consume disk. R21 is really in cache so the host is not able to access it. Needs more cache based on the amount of data transferred.
  4. R1 — S –> R21 — A –> R2 (Production site > Pass-thru Site > Out-of-region Site)
  5. Primary (R1) sites can have DMX-3 or DMX-4 or V-Max systems, Tertiary (R2) sites can have DMX-3 or DMX-4 or V-Max systems, while the Secondary (R21) sites needs to have a V-Max system.

R22 – Dual Secondary Devices:

  1. R22 devices can act as target devices for 2 x R1 devices
  2. One Source device can perform Read write on R22 devices
  3. RTO improved with primary site going down

Other Enhancements:

  1. Dynamic Cache Partitioning enhancements
  2. QoS for SRDF/S
  3. Concurrent writes
  4. Linear Scaling of I/O
  5. Response times equivalent across groups
  6. Virtual Provisioning supported with SRDF
  7. SRDF supports linking Virtual Provisioned device to another Virtual Provisioned device.
  8. Much more faster dynamic SRDF operations
  9. Much more faster failover and failback operations
  10. Much more faster SRDF sync’s

Some very limited V-Max Performance Stats related to SRDF:

  1. 36% improved FC performance
  2. FC I/O per channel up to 5000 IOPS
  3. GigE I/O per channel up to 4000 IOPS
  4. 260 MB/sec RA channel I/O rate, with DMX-4 it was 190 MB/seconds
  5. 90 MB/sec GigE channel I/O rate, with DMX-4 it was almost the same
  6. 36% improvement on SRDF Copy over FC
  7. New SRDF pairs can be created in 7 secs compared to 55 secs with previous generations
  8. Incremental establishes after splits happen in 3 seconds compared to 6 secs with previous generations
  9. Full SRDF establishes happen in 4 seconds compared to 55 seconds with previous generations
  10. Failback SRDF happen in 19 seconds compared to 47 seconds with previous generations

To read more about V-Max systems follow

To read more about SRDF systems follow