Google+

Archive

Author Archive

Expectations with new generation of DMX Technology

February 26th, 2009 No comments

There has been a big chatter about the next generation of EMC machines. After the initial release of DMX-3’s in 2005 and then the DMX-4’s in 2007, next generation DMX is almost due now.

With latest announcements from NetApp, IBM and EMC (Next Generation Celerra) in early Feb 2009, EMC’s DMX announcement might come right around EMC World 2009, plus or minus a month.

With the absence of Barry Burke from the Storage Blogosphere community over the past 3 months, it seems like he is busy working on strategy for the new generation of DMX machines.

Here is my wish list or expectations on the new DMX Platform.

Strategy, Cost, Marketing, Support…….

Do less with more!!!! This will have to be the reality of the new generation of DMX’s. In this tough economy and financial distress, if a new product is pitched with same efficiencies and overall similar ROI and TCO models, it will be hard to sell.

Some important CIO, CFO pitches would include, less foot print with more data, higher efficiency, delivered at 2/3rd the cost of previous generation of machines, energy savings, etc. Key Differentiators would be the cost per TB of data storage, cost per TB of management (Storage Administrators, OpEx), a savings of 20 to 30% in this equation might come into play.

The cost of warranty of each DMX is pretty high, during the warranty phase EMC Support Labs in Hopkinton, Sydney, Cork and Bangalore are supporting these boxes on a 24 x 7 basis. If EMC can manage to bring down the cost of in warranty repair including labor, parts, labs, engineering support, the savings from all these can be passed on to the customer. In this market, EMC might offer an extension of a 3 year warranty to a 4 year warranty that might help with ROI and TCO models.


Partners……

Support for new generation DMX’s installs extended to ASN Partners.

Some portions of Enterprise channels will be designed to work like Commercial channels promoting premier partners and ASN partners to perform some work on the enterprise machines.


Model Numbers……

The million dollar question, will it be called Symmetrix DMX-5?

Or will it be called
DMX-5-XP (Extra Performance),
DMX-5-EF (EFD optimized machine),
DMX-5-V (extended support for Virtualization),
DMX-5-950 (same naming convention as before),
DMX-5-8 (8GB I/O),
DMX-V (You can think its V for virtualization or V for roman letter 5)

Let’s not get hung up on the model numbers though.


Names…….

Will EMC for the first time drop the name Symmetrix from this generation of machines, this name comes from the Moshe days.

Technology…..

EMC is known to make a big bang with technology with all its new product releases and has been a leader in bringing new technology to the market. Let’s talk about a few technological aspects to look forward to in the new generation of DMX.

The underlying DMX-3 design has been different than the DMX and DMX2 generation of machines. The DMX-4 design has been prett
y similar to the DMX-3’s.

The point I am trying to make, EMC had a time frame of 5 years since the DMX2’s to come up with a radically changed DMX-5 design. Will a completely new design come to fruition with this generation of machines?

Enginuity Code……

A new Family Code is possibly due with this generation of DMX. May be an Initial release level of 5874.xx.xx.

Continued NDU (Non Disruptive Upgrade) Everything concept.

Introduction of PaPS (Plug and Play Support) with disk.


Size…….

2 Cabinet: where one Cabinet is for Controllers, and the other for 2.5 inch Flash Disk. This model will be optimized for Flash Drives only and will be lighting fast.

2 or more Cabinets where the 1st Cabinet is for Controllers and the other ones for drives, the additional drive cabinets can be used for 3.5 inch drives or for 2.5 inch drives depending on the cabinet type you purchase. Also supported with Flash Drives.

Total Raw Storage……..

2048 TB (Double the capacity from DMX-4)

Cache…..

1024 GB Cache (First Enterprise Storage Array to hit 1TB of Cache)

Maximum usable memory: 512GB

Controllers…….

Each DAF, GbE, FICON, ESCON controllers might be subdivided into 8 slices (ports, processors) creating further condensation of controllers, I/O, footprint, drives per DAF.

Additional backend ports will be added with this.

Each processor might be 2.4 Ghz PowerPC Chip.

Introduction of Clariion Type Concepts in Enterprise Storage…….

Plug and Play for disk replacements, where presence of an EMC CE onsite might not be required. This is pretty common with Clariion and NetApp Systems today.

Conceptual change of Global Memory to Local Memory, where memory is part of the controllers and not a global memory pool and Flash drives are used for certain memory operations as a vault.

Microcode upgrades being performed by the customer like its done for the Flarecode today.

EFD’s…….. b>

After support for 73GB and 146GB Flash, will might see 200GB , 400GB and 500GB disk on this new generation of machines?

With the use of EFD’s in Enterprise storage would the concept of IOPS with Storage change to GHz & MIPS.


Introduction of 2.5 inch drives…….

With some OEMs introducing support for 2.5 inch drives, we might see EMC moving in the same direction.

The next generation Clariion’s might have similar drives in them too.


Symapi…….

Today the Symapi database resides on the Service Processor. Service Processors die; crash or get interrupted in middle of a change (provisioning, allocation, and configuration) and all of sudden the customer finds themselves in middle of crisis. All the change windows scheduled will have to be rescheduled, PSE’s dialing into the boxes to troubleshoot and fix issues, etc.

Introduction of an IP based (ethernet) connection to the DMX (talking about the DMX and not the SP) with multiple paths of communication. The SYMAPI databases will be locally stored on the DMX rather than the SP. This is similar to the VCMDB and the SFS volumes which already reside on the DMX.

Introduction of VMware ESXi into the Service Processor Environment to run multiple instances of SP Software and Windows for diagnostics, remote call home, etc. May be One VMware install can call home on the highest priority errors to location 1 and the second calling home with low to medium errors at location 2 and create two different queues for support priority.

Ethernet…..

As mentioned earlier, an introduction of IP based Ethernet management port, allowing SMC (Symmetrix Management Console) to interface, ECC and other Components can communicate through the same infrastructure.

Hardware……

Channel support which would include FICON, ESCON, GbE, FC, iSCSI, RF and some initial support for FCoE.

2048TB of storage in 5 cabinets can only be achieved with 2.5 inch drives.

Introduction of LP SATA Drives…..

Introduction of Low Power SATA Drives to conserve energy.

Plug and Play Support……

It sounds unreasonable, but if this can be incorporated into the DMX Platform, it will really take the overall platform to new levels with configuration, provisioning, customer interface, management, etc. Imagine if you want to add new drives to your current DMX, no BIN file change, just plug the drives in and configure through SMC.

High Efficiency….

At least 30% increase in efficiency, usage, savings, power and reduction in administration, management, support.

Added interface friendliness for SMC usage.

I/O Improvements……

8GB I/O per second Backend?

Improved Cache Partitioning , Mirroring and Priority Controls……

Further enhancements related to cache partitioning and cache mirroring, allowing customers to prioritize cache based on applications, times of the day, etc to certain set of drives or interfaces.

New Conceptual Design with BCV’s, DRV’s, Snaps and Clones……

As I say conceptual, I am not sure if history can change with this new generation of DMX machines or the new code. The mirrors, bcv’s, drv’s, snaps, clones are all treated as mirrored positions, configuration like RAID-5, RAID-6 is hard to manage.

If the code has been completely rewritten using new technologies this might be a reality, working more at a lun level rather than drive levels.

EFD’s and Optimizer…….

With EFD’s the use of Symm Optimizer is not deemed necessary, will optimizer become history?

Policy based support for Atmos…….

This will be one of the best features to look forward with the release of new DMX’s. Will DMX have native support for Atmos or will it be through a policy based engine as additional physical hardware.

Enhanced Support for VMware……..

This is a given, limitations with DMX and VMware with usage of LUN #’s, with the new DMX we will see additional native support for VMware integration features.

Enhanced RSA Integration…..

A lot was seen with DMX-3 and DMX-4 with access controls, etc. Further enhancements to security aspects of the storage.

Enhanced Support for RAID 5 / RAID 6, possible modification of one of the RAID designs to make the product faster………

Introduction of a new RAID type with this generation of DMX’s to compete with NetApp’s RAID-DP.

Storage Virtualization…….

This is really questionable, not sure where EMC wants to take Storage Virtualization.

Virtual Provisioning licensing…….

Virtual Provisioning included as part of the microcode and at no additional cost.

Native support for Deduplication?

Could this happen with the latest DMX’s

Built in SRM tools?

Some support for build in SRM tools into SMC, will help customers identify issues with the DMX.

Advancements with Green Infrastructure……..

This is a given, big marketing pitch, energy savings of 30% at least.

Advancements in Workflow and Automation…….

Further advancements with Workflow, Automation in new versions of ECC and SMC.

Initial Support for FCoE…..

Initial FCoE support has been released on the Clariions. We will see some initial support for FCoE on the new DMX generation.

EMC has discussed this topic so many different ways in the blogosphere, I am pretty sure we will see some early adaption of it in this generation of DMX’s.


Hope I did cover a lot of ground in terms of new technology that we can look forward from EMC.

As usual comments always welcome.

Next Generation Celerra – Unified Storage with Deduplication – Feb 2009

February 23rd, 2009 No comments

After NetApp’s recent (February 2009) announcement of V-Series SSD support and IBM’s (February 2009) announcement of DS8000, EMC is on the roll next with the announcement of its Next Generation NAS product Celerra.

As usual, expected from EMC, the Big Bang!!!!!

So after a lot of speculation, finally the Next Generation Celerra is released now. Again this time around, EMC is pushing the technology towards unified storage, deduplication and virtual provisioning giving away some bells and whistles at no cost.

 

Here are the highlights of the product.

 

Celerra Next Generation Ultra Scale Architecture, Unified Storage with Deduplication, Virtual Provisioning, File Level Retention, Support for Flash Drives – 30X IOPS, LP SATA Drives 5.4K, 32% Energy Savings, 22% lower TCO, 960 drives, 960TB of RAW Storage.

 

Release date: 23rd Feb 2009

 

Product availability: Feb 2009, the NS-8G and NS-960 might be available early March 2009.

 

Models: NS-120, NS-480, NS-960, NS-G8 (Gateway Version).

 

Introduction of LP Sata Drives: Low Power SATA Drives 5.4K RPMs.

 

Introduction of Flash Drives in Celerra: 30X IOPS, introduction of Tier 0.

 

Cost: Low CapEx, OpEx. Customer installation available with Low and Medium profile celerra’s. High End Celerra’s available to install through ASN Partners or by EMC.

 

Protocols Supported: NAS, MPFS, FC, iSCSI

 

Software: Deduplication (no cost), Virtual Provisioning (no cost), Startup Assistant (no cost), Celerra Manager (no cost), Volume Manager (no cost), Celerra Snapsure (no cost) –

 

Energy Efficiency: 32% less energy consumption

 

Lower TCO: 22%

 

Build on: Intel Xeon Chips

 

Choice of Delivery: File Based or Block Based, NAS to MPFS for throughput, iSCSI to FC for throughput

 

NS-120

Supports 120 Drives

Supports Flash Drives

1 or 2 Blades

64TB

120TB RAW

 

NS-480

Support 480 Drives

Support Flash Drives

2 or 4 Blades

192TB

480TB RAW

 

NS-960

Support 960 Drives

Support Flash Drives

2 to 8 Blades

760TB

960TB RAW

 

NS-G8

Supports 4 Arrays behind NS-G8

2 to 8 Blades

896TB RAW

 

Applications usable on Celerra: VMware, Oracle, MS Exchange, MS SQL Server, Windows, Linux File Server

 

Celerra Integration Available: With VMware, Oracle, MS Exchange, MS SQL

 

Classifications:

High End: NS-G8, NS-960

Mid-Tier: NS-40G, NS-480, NS-120

Low End: NX-4

 

Compliance: Meets file level compliance related to SEC Rule 17a-4(f). Also available for the Celerra is 3rd Party Compliance.

 

Celerra File Level Retention: Celerra is being pushed to allow Filesystem archiving. For Application and Filesystem archiving you will still need a Centera.

 

ROI Models: Better ROI on Celerra models than any comparative NetApp  models.

Above are the product highlights, a technical blog on this coming soon……

Haven’t had a chance to play with it yet, but hopefully soon and looking forward to it. 

Edit: Read Dave Graham’s Blog Post on Celerra Here……

FeedBurner Issues – Size limit – 512KB

February 17th, 2009 2 comments

Created a blog post about a month ago in which I discussed how excited I was to use pictures on all the blog post going forward. 

it worked good………

On Friday, got a twit from one of the followers complaining that podcast through Odigo wasn’t working on the last 3 post. The blog post were created and uploaded to the blog site about 3 to 6 days ago.

So started looking around for the problem, thinking it was an Odigo problem…….

Upon further research realized FeedBurner wasn’t working correctly. No new posts were been delivered to any RSS feed readers including Odigo Feed Fetcher.

Panic mode…..

After poking around for an hour and reading some tech articles, realized the feed size cannot be larger than 512kb (what is 512kb in this day and age). Apparently all the pictures that I had been posting on the blog were consuming part of the 512kb max feed size. So that is it, no more feeds being delivered to any readers…..

Didn’t realize there was a FeedBurner size limit, are YOU aware of it……….

Lesson learned….

Be careful putting up your feeds on FeedBurner, Size limit is 512kb and your pictures count in it. Check every week to see if your feeds are working okay. I believe your videos would count as part of the limit as well.

Look and explore other possible options……..

So now I subscribe to the health of the feed through RSS reader. Also limited the feed results by changing the initial feed address, limiting a reader to only download the last 20 posts. Once this was added to the original feed address, all started working again.

Don’t post too many pictures on your blog if you are using FeedBurner…….

SUN StorageTek’s RAID 6 Implementation

February 16th, 2009 No comments

Continuing the series on RAID 6 technology with different OEM products; this time around it’s SUN StorageTek version of RAID 6 implementation on 6xxx series arrays.

Here are the links to previous posts related to RAID 6. It covers NetApp’s RAID-DP, Hitachi’s RAID 6, HP’s RAID 6 and technical post on various different RAID technologies.

HP’s RAID 6

NetApp’s RAID–DP

Hitachi’s (HDS) RAID 6

Different RAID Technologies (Detailed)

Different RAID Types

SUN’s Business Case

Similar to HP, Sun is not making a claim on performance with RAID 6; its only claim with this technology implementation on its products is Improved Reliability.

RAID 6 is supported on SUN StorageTek’s 6140, 6580 and 6780 Arrays

SUN StorageTek products traditionally support RAID 0, RAID 1, RAID 3, RAID 5, RAID 1+0 and now RAID 6.

Different Applications that can use RAID 6 technology on SUN StorageTek arrays are Exchange Email, Web Servers, File Servers, Oracle, SAP, SQL and MySQL.

Technology

Here is a small abstract about RAID 6 Technology as discussed in my earlier post.

Technology: Striping Data with Double Parity, Independent Data Disk with Double Parity

Performance: Medium

Overhead: 10% to 50% overhead, with additional drives you can bring down the overhead.

Data Loss: With one drive failure and two drive failures in the same Raid Group no data loss.

Advantages: RAID 6 is essentially an extension of RAID 5 which allows for additional fault tolerance by using a second independent distributed parity scheme (two-dimensional parity). Data is striped on a block level across a set of drives, just like in RAID 5, and a second set of parity is calculated and written across all the drives; RAID 6 provides for an extremely high data fault tolerance and can sustain multiple simultaneous drive failures which typically makes it a perfect solution for mission critical applications.

Disadvantages: Poor Write performance in addition to requiring N+2 drives to implement because of P+Q parity scheme.

The technology (P+Q Parity Algorithm) of Sun’s implementation and HP’s implementation of RAID 6 is similar. To further read about HP’s RAID 6, please use the link below.

HP’s RAID 6

The formula to calculate your usable space with RAID 6 is C * (n – 2), where C is the Size of your smallest drive in the RAID Group and n = number of drives. It is highly recommended all your disk drive sizes are similar.

The largest configuration of RAID 6 can be 257 drives with 2 parity drives and 255 data drives. Usage of more drives will help bring down the RAID 6 overhead since 2 drives are always allocated as parity.

This technology can be implemented with a minimum of 4 drives. The overhead with use of 4 drives in a single RAID Group is 50%. If you run 60 drives in a single RAID group, your overhead might be close to 4% {100 – [100 x 2 (parity) / 60 (drives)]}.

Lower write performance is a unique disadvantage of using RAID 6.

Both Sun StorageTek’s RAID 6 and HP’s RAID 6 offers advantages related to reliability and data protection but no additional benefits with performance, capacity, efficiency, speed, etc.

We will leave it up to the user to decide what RAID technology best fits their environment on SUN StorageTek 6XXX arrays for various different applications.