RAID Cloud: A New IT Architecture

Rick Parker, IT Director, Fetch Technologies | 11/29/2011 | 15 comments

Rick Parker
The current IT architecture process is to order the number of racks needed to support initial requirements, order some additional empty racks with a Right of First Refusal, and add a backup datacenter when the company has enough resources.

The problem with this process is that the end result is usually not a cost-effective design. The datacenter just continues to grow ad hoc as projects are added, becoming less cost-efficient, more complicated, less reliable, and more difficult to manage over time. Most companies never get big enough to afford a backup datacenter and 100 percent redundancy cost -- and if they do, the backup datacenter is built to a completely different design than the primary one, so it's useless to test one against the other.

RAID Cloud Architecture (RCA) is a revolutionary IT architecture -- a collection of one or more datacenters -- based on the same proven design as Disk RAID systems. The major difference between current IT and RCA is that RCA is not built in an ad hoc way; it is an infinitely scalable design starting with a minimum of three hybrid cloud datacenters for a maximum 33 percent redundancy cost.

A RAID cloud architecture design starts with a combination of three or more datacenter modules (DM). A DM is a complete configuration of routers, switches, servers, storage, etc. DMs are connected at Layer Two (of the OSI Seven Layer Model) to create a single logical architecture.

In a six DM design, the first two DMs should be located very close together and close to the main Enterprise office for low-cost point-to-point connectivity from the office and to each other, and to create the first datacenter cluster for RAID 1c redundancy. Assuming that, for budgetary reasons, they need to be built as needed, the third datacenter should be in the center of the US, the fourth on east coast of the US, the fifth in the center of the US, and the sixth in the eastern US to create three Datacenter Clusters with an overall 16 percent redundancy cost. The Datacenter Clusters are created to support live vMotion and storage replication for 100 percent availability.

RCA is designed to be cost effective due to higher virtualized resource and datacenter space utilization and the use of smaller, less expensive components than those used in a "traditional" datacenter. Using multiple datacenter co-location and bandwidth vendors for each datacenter module provides leverage for cost negotiation and price protection. The modules are a small cookie cutter designed to be inexpensive.

RCA is designed for infinite 100 percent reliability by using many multiple colocation and bandwidth sources. Using global server load balancing, an entire datacenter can be down for maintenance or service failure and only cause a loss of capacity, not availability. RCA is scalable: As more capacity is needed, more DMs are added until six DMs are created. As more capacity is needed beyond six, each DM can be quickly increased in size. Because DMs are generally built as needed, they are contracted over time, so if less capacity is needed, DMs can be decommissioned. This flexibility makes RCA a very efficiency design.

DMs are designed to be quickly configured, usually in under a week, and all the components are the same, so the configurations can be duplicated. Multiple duplicate DMs allow testing of new versions of system software without affecting an entire architecture. Multiple DMs also provide swap space to upgrade systems; in a single datacenter, space has to be left open to install new systems or significant downtime has to be scheduled to support in place upgrades.

RCA is based on a heavily virtualized infrastructure so capacity can be quickly added. In addition to low-cost connections, DMs are located close together and close to the office, so multiple DMs can be visited with minimum travel expense incurred by technical staff. One company estimates a $20k+ annual savings in driving time alone for the LA area.

Taken as a whole, the RCA design offers significant advantages over traditional datacenters, especially if redundancy is a primary concern. If the IT efficiency arguments don't get your attention, perhaps the savings in gasoline and employee miles can draw your interest.

View Comments: Newest First | Oldest First | Threaded View
Page 1 / 2   >   >>
@parkercloud   RAID Cloud: A New IT Architecture   12/12/2011 1:07:31 AM
Re: Lots of benefits here
There are even more benefits that I did not mention. Due to 100% virtualization remote console systems are not needed (supplied by hypervisor), monitoring is less expensive, much less hardware maintenance because much fewer physical devices, much better remote access because servers are in data centers with faster / chaper server connections then office, faster server maintenance / management again because faster / cheaper internet  connections for faster patch / application downloads  and installs. Much more reliable because much fewer hard drives and redundant drive contollers in storage systems
@parkercloud   RAID Cloud: A New IT Architecture   12/12/2011 1:00:43 AM
Re: Savings
This design is meant for existing networks as well as for new networks. This design works because it is not dependant on new technology. Cloud Architecture is primary configuration more then a technology and all the technologies are standard and proven and most enerprises have staff familair with these technologies.

Existing systems can be reconfigured as needed. They might not be the best option but they can work until they are expensed, support expires, end of life etc. The primary technology that usually needs to be added is virtualization and a Cloud Management Application like Platform ISF. Abiquo, Embotics, etc. 
@parkercloud   RAID Cloud: A New IT Architecture   12/11/2011 3:32:41 PM
Re: Would this work for resource poor businesses?
The initial outlay varies on the size of the individual datacenters, I have designed an entire datacenter in a single rack including all components; servers, SAN storage, backup, remote access, monitoring, etc. I expect most organizations will use from 1 to 5 racks per datacenter. Its possible and expected to start with a single data center if a single rack datacenter figure app. $150k that could vary a lot depending on hardware / initial capacity selected. Leasing is also an option to get a monthly cost as opposed to a single large cost.
@parkercloud   RAID Cloud: A New IT Architecture   12/11/2011 12:51:26 PM
Re: Lots of benefits here
There are additional bandwidth costs for connections between datacenters but I expect the cost savings of Cloud / Virtualization to more then cover that. There is also significant cost savings in the specific hardware / models used in this Architecture. The challenge, the huge almost impossible challenge is a human challenge not a technical challenge. To do this requires significant change and cooperation between usually siloed groups, ie. storage, networking, etc. I doubt most groups are willing to make the change and most IT executives willing to force change. 
@parkercloud   RAID Cloud: A New IT Architecture   12/11/2011 12:29:29 PM
Re: Savings
I think there are partial redundancy / backup options but I have only heard of 2 location multi-site offers or multi-site connections or multi-site backups but not multi-site full IT duplication 
@parkercloud   RAID Cloud: A New IT Architecture   12/11/2011 12:24:06 PM
Re: Savings
I would especially use this plan if I was starting to build new data centers. Because this isn't just a data center plan it is an Architecture to support the entire Enterprise for a significant amount of time. Because it is a Hybrid Cloud Architecture it should involve reorganization of how IT is managed to drive IT into the future. Why use this Plan? Only if it is better then alternatives, I think the primary alternative is the classic primary / backup plan and I hope I have explained why RAID Cloud is better then that.
@parkercloud   RAID Cloud: A New IT Architecture   12/11/2011 12:08:48 PM
Re: RAID Cloud: A New IT Architecture
In a RAID Cloud Architecture there isn't a Primary or Backup datacenter they are all equal
batye   RAID Cloud: A New IT Architecture   12/1/2011 11:17:34 AM
Re: Lots of benefits here
same question here as per Sara question- Where is the catch?

kstaron   RAID Cloud: A New IT Architecture   11/30/2011 12:22:37 PM
Would this work for resource poor businesses?
If I were a business that needed redundancy but has very little resources to put toward it, Would this be the way to go? Most things start ad hoc because it starts off as a couple of guys in a basement, or as an after thought. If from the very beginning we want datacenters such as you describe, what is the initial outlay?
tekedge   RAID Cloud: A New IT Architecture   11/30/2011 12:16:31 PM
RAID Cloud: A New IT Architecture
" the backup datacenter is built to a completely different design than the primary one, so it's useless to test one against the other"

I  have a feeling that with the RAID cloud architecture in place we will be still seeing quite some  differences between the primary and the backup data centers which means it wd still need some work  to test one against the other.
Page 1 / 2   >   >>

The blogs and comments posted on do not reflect the views of TechWeb,, or its sponsors., TechWeb, and its sponsors do not assume responsibility for any comments, claims, or opinions made by authors and bloggers. They are no substitute for your own research and should not be relied upon for trading or any other purpose.

More Blogs from Rick Parker
Rick Parker   8/5/2011   11 comments
Cloud computing, especially private/hybrid cloud computing, consists of shared hardware, so monitoring becomes much more critical. If a performance or capacity limit is reached, multiple ...
Rick Parker   12/16/2010   11 comments
Private cloud management is different from current network management. In a private cloud, end user business staff is responsible for managing their own IT resources, and IT is responsible ...
Rick Parker   12/9/2010   21 comments
I have been reading and hearing a lot of bad information and incorrect opinions over the past few months by experienced IT staff and self-proclaimed cloud computing experts: people ...
Rick Parker   10/26/2010   15 comments
If I ever buy any more server-based storage (and I don't plan to), it will be in very small increments. It's expensive, unreliable, and inefficient. And with private clouds on the horizon, ...
Latest Archived Broadcast
In this episode, you'll learn how to stretch the limits of your private cloud -- and how to recognize the limits that can't be exceeded.
On-demand Video with Chat
IT has to deploy Server 2012 in a way that fits the architecture of its application delivery system.
E2 IT Migration Zones
IT Migration Zone - UK
Why PowerShell Is Important
Reduce the Windows 8 Footprint for VDI
Rethinking Storage Management
IT Migration Zone - FR
SQL Server : 240 To de mémoire flash pour votre data warehouse
Quand Office vient booster les revenus Cloud et Android de Microsoft
Windows Phone : Nokia veut davantage d'applications (et les utilisateurs aussi)
IT Migration Zone - DE
Cloud Computing: Warum Unternehmen trotz NSA auf die „private“ Wolke setzen sollten
Cloud Computing bleibt Wachstumsmarkt – Windows Azure ist Vorreiter
Like Us on Facebook
Twitter Feed
Enterprise Efficiency Twitter Feed
Site Moderators Wanted
Enterprise Efficiency is looking for engaged readers to moderate the message boards on this site. Engage in high-IQ conversations with IT industry leaders; earn kudos and perks. Interested? E-mail:
Informed CIO: Dollars & Sense: Virtual Desktop Infrastructure
Cut through the VDI hype and get the full picture -- including ROI and the impact on your Data Center -- to make an informed decision about your virtual desktop infrastructure deployments.

Read the full report
Virtualization Management: Time To Get Serious
Welcome to the backside of the virtualization wave. Discover the state of virtualization management and where analysts are predicting it is heading

Read the full report
A Video Case Study – Translational Genomics Research Institute
e2 Storage Video

On the Case
TGen IT: Where We're Going Next

7|11|12   |   08:12   |   10 comments

Now that TGen has broken new ground in genomic research by using Dell's storage, cloud, and high-performance computing solutions, the company discusses what will come next for it and for personalized medicine.
On the Case
Better Care Through Better Communications

6|6|12   |   02:24   |   11 comments

The achievements of the TGen/Dell project could improve how all people receive healthcare, because they are creating ways to improve end-to-end communication of medical data.
On the Case
TGen IT: Where We Are Now

5|15|12   |   06:58   |   6 comments

TGen is breaking new ground in genomic research by using Dell's storage, cloud, and high-performance computing solutions.
On the Case
TGen IT: Where We Were

4|27|12   |   06:45   |   10 comments

The Translational Genomics Research Institute wanted to save lives, but its efforts were hobbled by immense computing challenges related to collecting, processing, sharing, and storing enormous amounts of data.
On the Case
1,200% Faster

4|18|12   |   02:27   |   12 comments

Through their partnership, Dell and TGen have increased the speed of TGen’s medical research by 1,200 percent.
On the Case
IT May Improve Children's Chances of Survival

4|17|12   |   02:12   |   8 comments

IT is helping medical researchers reach breakthroughs in a way and pace never seen before.
On the Case
Medical Advances in the Cloud

4|10|12   |   1:25   |   5 comments

TGen and Dell are pushing the boundaries of computing, and harnessing the power of the cloud to improve healthcare.
On the Case
TGen: Living the Mission

4|9|12   |   2:25   |   3 comments

TGen's CIO puts the organizational mission at the heart of everything the IT staff does.
On the Case
TGen Speeding Up Biomedical Research to Save More Lives

4|5|12   |   1:59   |   6 comments

The Translational Genomics Research Institute is revamping its computing to improve speed, storage, and collaboration – and, most importantly, to save lives.
On the Case
Computing Power Helping to Save Children's Lives

3|28|12   |   2:13   |   3 comments

The Translational Genomics Institute’s partnership with Dell is enabling them to treat kids with neuroblastoma more quickly and save more lives.
Tom Nolle
How Deep Is My Storage Hierarchy?

7|3|12   |   2:13   |   5 comments

At the GigaOM Structure conference, a startup announced a cloud and virtualization storage optimizing approach that shows there's still a lot of thinking to be done on the way storage joins the virtual world.
E2 Interview
What Other Industries Can Learn From Financial Services

6|13|12   |   02:08   |   3 comments

We asked CIO Steve Rubinow what CIOs in other industries can learn from the financial services industry about datacenter efficiency, security, and green computing.
E2 Interview
Removing Big-Data Flow Bottlenecks

6|12|12   |   02:55   |   No comments

We ask CIO Steve Rubinow what pieces of financial services infrastructure need to perform better to get traders info faster.
E2 Interview
Getting Traders the Data They Need

6|11|12   |   02:04   |   1 comment

We ask CIO Steve Rubinow: What do stock market traders need to know, how fast do they need it, and how can CIOs get it to them?
E2 Interview
Can IT Help Fix the Global Economy?

6|8|12   |   02:32   |   2 comments

We ask CIO Steve Rubinow whether today's IT can help repair the global economy (and if IT played any role in the economy's collapse).
E2 Interview
More Competitive Business via Datacenter Strategy

5|4|12   |   2:46   |   1 comment

Businesses need to be competitive, yet efficient, and both goals affect datacenter design.
E2 Interview
The Recipe for Greater Efficiency

5|3|12   |   3:14   |   2 comments

Intel supplies the best ingredients to drive greater datacenter efficiency and support new compute, storage, and networking needs.
E2 Interview
Datacenters Enabling Business Transformation

5|1|12   |   06:37   |   1 comment

Dell’s Gaurav Chand says that for the first time ever datacenter technology is truly enabling all kinds of organizations to transform their business and achieve new objectives.
Tom Nolle
Cloud Data: Big AND Persistent!

3|28|12   |   2:11   |   10 comments

We always hear about "Big" data, but a real issue in cloud storage is not just bigness but also persistence. A large data model is less complicated than a big application repository that somehow needs to be accessed. The Hadoop send-program-to-data model may be the answer.
Tom Nolle
Project Lightning Streamlines Storage

2|16|12   |   2:09   |   2 comments

EMC's Project Lightning has matured into a product set, and it's important, less because it has new features or capabilities in storage technology and management, than because it may package the state of the art in a way more businesses can deploy.
Tom Nolle
Big Data Appliance Is Big News

1|12|12   |   2:18   |   No comments

Oracle's release of a Hadoop appliance for Big Data may be a signal that we're shifting to database appliances.
Tom Nolle
Myopia Can Hurt Storage Policy

12|22|11   |   2:08   |   No comments

We're at the beginning of a cloud-driven revolution in storage, but Oracle's quarter shows that enterprises are hunkering down on old concepts because they're afraid of the costs in the near term.
Sara Peters
An Untrained User & a Mobile Medical Device

12|19|11   |   2:43   |   11 comments

Untrained end users, clueless central IT staff, and expensive mobile devices are a worrisome combination for healthcare CIOs.
Tom Nolle
Too Many Labels on 'Big Data'?

12|9|11   |   2:12   |   3 comments

However you label it, structured and unstructured information are different and will likely always require different tools.
Sara Peters
E2 Debuts New Storage Section

12|8|11   |   1:51   |   1 comment

Need strategic guidance on everything from SSDs to 100 percent virtualized datacenters? Look no further.