Cyber Infrastructure org goes with OpenStack and Dell

June 20, 2011

Cybera, a Canadian not-for-profit recently selected OpenStack along with Dell systems to build out their Infrastructure as a Service cloud.  The organization, which is based in Alberta, “collaborates with public and private sector partners to accelerate research and product development that meets the needs of today’s society.”

Most recently Cybera used OpenStack to build out a cloud for CANARIE’s (Canada’s Advanced Research and Innovation Network) DAIR project.

Here’s what Cybera had to say in their blog:

To start with, you’ll need hardware. If you have the time and inclination, the best thing to do might be to ask Rackspace Cloud Builders for some help spec’ing out the hardware for OpenStack. This is the route that Cybera went and we got some badly needed advice. Since you might not be able to go that route, I’ll tell you what we know.

At the end of the day we went with Dell, based on the Cloud Builders’ advice and our own due diligence. If you aren’t aware of it yet, Dell is supporting OpenStack in a big way. They have a number of pages dedicated to it here. There’s also a whitepaper that discusses hardware and network for OpenStack, if you feel like filling out the form.

We ordered four different types of servers (aka nodes). A management node (nova-api, nova-network, nova-scheduler, nova-objectstore), compute nodes (nova-compute, nova-volume), a proxy node (swift-proxy-server) and storage nodes (swift-object-*, swift-container-*, swift-account-*). All nodes were contained in the Dell C6100 chassis. Here are the specs:

Processor Sockets Cores Threads RAM Disk
Management E5620 2 8 16 24 8 x 300 GB
Compute X5650 2 12 24 96 6 x 500 GB
Proxy E5620 2 8 16 24 4 x 300 GB
Storage E5620 2 8 16 24 6 x 2 TB

Great to see people picking up OpenStack and running with it!

Read more about Cybera’s experience at Running OpenStack in Production: Part 1: Hardware

Pau for now…


TIBCO, Dell, Arista & Mellanox deliver Extreme Low-Latency messaging solution for Financial Markets

April 19, 2011

When it comes to latency in financial trading, the name of the game is how low can you go.  According to an industry analyst, a millisecond of latency in trading can translate into $100 million of lost revenue for a major brokerage firm.

In a quest to quash latency while at the same time providing flexibility and agility, TIBCO has banded together with Dell, Arista Networks and Mellanox Technologies to deliver a complete Extreme-Low-Latency Messaging solution based on TIBCO FTL.

This solution, which was announced earlier today, is made up of the following:

  • TIBCO FTL
  • Arista 7100 switches
  • Dell PowerEdge C6100 servers
  • Mellanox ConnectX-2 EN 10GigE NICs with RDMA over Converged Ethernet

This end-to-end solution enables financial trading market customers to rapidly turn millions of messages per second into actionable information on the trading room floor.  Faster access to information means faster time to analysis and faster time to profitability.  And in today’s global financial markets, speed is king.

Extra-credit reading

  • Datasheet: TIBCO FTL: Extreme Low-Latency Messaging
  • Press release: TIBCO, Arista Networks, Dell, and Mellanox Technologies Team Up to Provide Financial Markets with Extreme Low-Latency Messaging Solution

Pau for now…


Dell DCS unveils its 4th HPC offering in 12 months, and its a beefy one

February 14, 2011

Today Dell Data Center Solutions (DCS) is announcing the PowerEdge C6145, number four in our line of offerings targeted specifically at High Performance Computing.  This AMD-based system, which contains two four-socket servers for a total of 96 cores, ranked as the highest performing x86 2U shared infrastructure server on the market based on SPECfp_rate2006 results. In addition, the PowerEdge C6145 can deliver up to a 534% better price performance at 1/5 the cost and 1/4 of the rack space when compared to HP’s ProLiant DL980 G71.

The HPC beat goes on

When we in DCS launched our PowerEdge C line almost a year ago, our first HPC-focused machine was the Intel-based C6100.   We followed it three months later with our C410x expansion chassis to supercharge it and then, three months after that, we came out with the AMD version of the C6100, the PowerEdge C6105.   Now three months after that system debuted we are unveiling the C6145.  All three servers come in the same 2U package but with differing chips and architectures targeted at different HPC application types.

Check out the video below and let the C6145 architect, John Stuewe take you on a quick tour of this new muscle machine.

Hairy problem solver

The PowerEdge C6145 with its 755FLOPS and up to 1T of memory is specifically geared to solving big problems involving huge and complex data sets in mapping, visualization, simulations and rendering, and solving them faster.  With regards to efficiency, the shared infrastructure design of the system can reduce the number of individual fans by 1/4 compared to traditional 2U systems with less power needed to cool and resulting in higher performance per watt, per dollar.

Super charge it

As if 96 cores packed into 2U wasn’t powerful enough, you can take your workloads “to 11″ with the help of the PowerEdge C410x.  The C410x PCIe expansion chassis allows you to double the server to graphics processing unit (GPU) ratio to 1:8 to create a number-crunching uber powerhouse.

Dell DCS has been listening to their HPC customers and rolling out systems to meet their needs, today we’ve announced the latest in our line up, the PowerEdge C6145.

Extra-Credit reading:

Pau for now…

1 Based on testing by Dell Labs. Dell PowerEdge C6145: SPECfp_rate2006 of 1310 in 2U as compared to HP ProLiant DL980 G7: SPECfp_rate2006 of 1080 in 8U.  SPEC® and the benchmark name SPECfp® are registered trademarks of the Standard Performance Evaluation Corporation.  Competitive benchmarks stated above reflect results published or submitted to www.spec.org as of Feb 14, 2011.  The comparison presented above is based on the best performing 8-chip x86 servers.  For the latest SPECfp_rate2006 benchmark results, visit http://www.spec.org/cpu2006. Actual performance will vary based on configuration, usage and manufacturing variability.

Dell provides Ubuntu-powered IaaS-in-a-box

February 3, 2011

Yesterday, the announcement went out that the Dell | Canonical Enterprise Cloud, Standard Edition was out and ready for consumption.  What this cloud-in-a-box allows folks to do is to set-up affordable Infrastructure-as-a-Service (Iaas)-style private clouds in their computer labs or data centers.  The cool thing is that, because the Ubuntu Enterprise Cloud (UEC) software  is compatible with Amazon Web Services EC2 and S3 services, it enables IT admins and developers to move workloads between public and private clouds.

Who cares?

Application developers and IT service providers and admins who are setting up cloud POC’s are perfect candidates for this pre-configured testing and development environment.  With regards to industries, areas where there is a lot of software development work like Hosters, Telco & Communications, Media & Entertainment and Web 2.0 businesses are prime markets for the Dell UEC solution.

So what’s in it?

The solutions’ basic components are Dell PowerEdge C systems plus a Dell-specific download of the Ubuntu Enterprise Cloud (made up of the Ubuntu operating system and the Eucalyptus platform for private cloud computing).  To simplify getting the whole shebang up and running Dell and Canonical are providing the following:

Here’s a peak at the hardware that supports it:

The Dell UEC cloud solution pod.

  • Cloud Compute Server – PowerEdge C6100 that embeds four discrete compute nodes in a single enclosure
  • Cloud Front-end Server – PowerEdge C2100 server that acts as an all-in-controller and runs all shared UEC-related services
  • Infrastructure Server – PowerEdge C2100 that runs two components of the cloud infrastructure:
    • Cloud Deployment and Landscape Management
    • Cloud Storage
  • Network switch – PowerConnect 6248

And on the software side…

The architecture looks something like this:

The Dell UEC cloud solution architecture

The software components are:

  • Cloud Controller (CLC) – the cloud portal
  • Walrus Controller (W) – the cloud’s storage repository
  • Cluster Controller (CC) – the controller for a up to 1024 compute cores grouped together as a cluster
  • Storage Controller (SC) – the controller for cluster’s storage repository
  • Compute Node (CN) – cloud’s compute node

And on the support side…

If you’re looking for systems management and support services with your order, you are in luck.  Canonical, the company behind Ubuntu, has put together UEC Assist, a support service built specifically for Dell customers deploying SE Edition and which is delivered by Canonical’s Global Services and Support team.

Its all about efficiency

From a Dell DCS (the group at Dell behind this) point of view, this offering fits in well with our strategy of bringing total solutions to market that optimize efficiency at every layer, from code to servers to storage.  The open source Dell UEC solution is tailor made to deliver a ready to go IaaS solution.

Extra-credit reading:

Pau for now..


Mark Shuttleworth on UEC and OpenStack

January 11, 2011

Mark Shuttleworth, the ever gracious founder of Ubuntu, stopped by Dell this morning to talk to various folks about various subjects.   I was able to grab some time with him between meetings and get his thoughts on a few topics.

I was particularly interested in getting his thoughts on Ubuntu Enterprise Cloud (UEC) which will be available this week from Dell as the Dell | Canonical UEC Solution (along with the UEC software, the solution is based on our PowerEdge C2100 and C6100 and comes with a reference architecture and deployment guide).   The other topic I wanted to get his insight into was OpenStack.

Here’s what Mark had to say:

What Mark talked about

  • How Mark has settled into his role as non-CEO (he is still chairman).
  • What he is focusing on these days: the cloud and product design e.g. Unity.
  • [1:45] The thinking behind UEC and the combined Dell Canonical offering.
  • [3:45] OpenStack and Canonical’s participation
  • Working with both OpenStack and Eucalyptus and how both of these are central to the process of standardization that we are starting to see at the infrastructure layer of cloud computing.

Extra-credit reading:

Pau for now…


What’s happening — A Dell / OpenStack Update

September 15, 2010

A couple of days ago Bret Piatt, who handles Technical Alliances for OpenStack, came up to Austin to have further discussion with our team’s software engineers around OpenStack.  If you’re not familiar with OpenStack, it is an open source cloud platform founded on contributed code from Rackspace and NASA’s Nebula cloud.

The project was kicked off a couple of months ago at an inaugural design summit held here in Austin.  The summit drew over 25 companies from around the world, including Dell, to give input on the project and collectively map out the design for the project’s two main efforts, Cloud Compute and Object Storage.

Since the summit, and the project’s subsequent announcement the following week at the OSCON Cloud Summit, the community has been digging in.  The first object storage code release will be available this month and the initial compute release, dubbed the “Austin” release, is slated for October  21.  Additionally, the second OpenStack Design Summit has been set for November 9-12 in San Antonio, Texas, and is open to the public.

OpenStack visits Dell

During Bret’s visit to Dell he met with a bunch of folks including two of our software architects, Greg Althaus and Rob Hirschfeld.  The three talked about how things were going with the project since the summit as well as specific ways in which Dell can contribute to the OpenStack project.

Below you can see where I crashed the three’s whiteboard session and made them tell me what they were doing.  I then followed them, camera in hand, down to the lab where Greg and Rob showed Bret the system that we have targeted for running OpenStack.

Some of the topics (L -> R) Bret, Greg and Rob touch on:

  • Bret:  Getting ready for the object storage release in September and compute in October.  Looking to get the right hardware spec’d out so that people can start using the solution once its released.
  • Rob: Learning about how the project is coming together since the design summit.  Interested in how the 3 code lines, storage, NASA compute and Rackspace compute, along with the input that was gathered at the Design summit and community input, are coming together.
  • Greg and Rob take Brett to the lab to show him the C6100 which could be a good candidate for open stack.
  • Next step, getting OpenStack in the lab and start playing with it.

Extra-credit reading:

Pau for now…


Follow

Get every new post delivered to your Inbox.

Join 83 other followers

%d bloggers like this: