Blog

Infrastructure-as-a-Service Breakdown

There’s an interesting post by Advanced Systems Group that provides the top-level costs they were able to derive from an internal study they did to compare public and private cloud implementations.  It’s the start of a good discussion (questions/comments I posted are still in moderation).

Their post compares three different cloud options from a cost perspective: 1) Public Cloud from Amazon EC2; 2) Purchased Private Cloud infrastructure; and 3) Leased private cloud infrastructure.

All three systems were based around the equivalent of (using AWS parlance) 10 Extra Large CPU instances, 60 Large CPU instances, and 30 Small CPU instances.

The cost calculations were run by ASG over 3 years assuming an increase of 15% in compute demand, and not accounting for depreciation.

When ASG noted in an answer to a comment that all instances are assumed to be running 24/7 my curiosity was piqued to drill in a little bit.  The problem is, solely comparing private and public based upon 100% utilization removes one of the most compelling reason for the cloud - a pay-as-you-use model.  It skews the conclusion from the start.  (How many data centers run at 100% utilization anyway?)

Also, from a rough calculation, we can see that their Amazon numbers appeared to be based upon the cost of On-Demand Instances.  24/7 utilization over a three year period is not a cost effective use model for the On-Demand pricing model.

Performing a straight hardware costs calculation, (I’m assuming that ASG were adding in some additional costs for storage and data transfers to get to their top line numbers, as I couldn’t get there from a simple calculation.  Their 3Yr cumulative cost for the Public Cloud scenario was slightly over $1M, where as mine below came in slightly under), I got the following top line numbers for an On-Demand instance scenario (the least economical).  For completeness, I also added in the two more sensible scenarios of either a rolling 1 Year Reserved Instance model or a one-time 3 Year Instance model.

Public vs Private Cloud Costs (100% Utilization)

And so, taking these basic calculations in isolation of all other considerations, we can see that from a purely hardware resource expenditure perspective, a Public Cloud implementation is roughly the same or even less than a private cloud implementation.

However, this is still only part of the story.  As mentioned, a key draw to public cloud computing is the elasticity of the model - Pay-as-you-Use.  And so, considering a selection of rough percentage utilizations, we get the following breakdown of the different hardware costs of a public cloud implementation.

AWS Public Cloud Costs: Time vs Model vs Utilization

The comparison relative to utilization becomes even more stark when charted.

Cloud Cost Comparison vs Percentage Utilization

Not only is AWS more economical over utilization rates over extended periods, but it provides flexibility.  For example, if the software used on the servers is upgraded (as is likely) and hardware requirements increase (equally likely, given history), moving an image (AMI) from a Small to a Large instance is straight forward (for an “IT guy”).  No hardware needs to be thrown out or shipped in.

Additionally, there is some flexibility around migrating across business models.  Of course, it is easier to upgrade than downgrade

Posted under cloud

It will be interesting to see what the corporate landscape will look like when we get to the other side of the recession.  According to a report by The McKinsey Quarterly, it will be quite different to how it looks now.

“[After analyzing] nearly 700 high-tech companies during contractions in markets around the world over the past two decades. [They] found that the turmoil accompanying downturns significantly reconfigures the high-tech landscape. About half of the companies that entered these downturns as leaders—the top 20 percent—ended up as laggards when the economy regained momentum.”

There are many ways that this could come about of course, some are:

1. The leaders were slow to react to changing economic pressures and costs dragged them down
2. The leaders saved cash and did not invest in technology to keep up with market demands
3. Entrants or non-leaders innovated around the leaders with a business model that the downturn actually strengthened

Of course, “business model” is not limited to “pricing” or “cost structure”.  It could include the channel, licensing model, support structure, etc.  Everything that goes towards how a company makes and uses money.

In a software-based industry, there is a great deal of opportunity in #3.  And if you’re interested in discussing how Xuropa can help your company take advantage of the environment, please drop me a line via Twitter (sfojames), LinkedIn, or Xuropa.

Posted under Xuropa, business, industry

An alternative title could be: “Opportunities for the Next Wave in Electronic Design”

Hopefully you caught my last two posts recapping a conversation with Gary Smith.  If not, here they are:

A Breath of Fresh Air: Lunch with Gary Smith, Part 1

Lunch with Gary Smith, Part 2

Having laid the ground work, the next topic of conversation was the most exciting.  We are truly at an important juncture in the industry and this is how Gary described it:

Opportunity #1: Multi-core Support by Embedded Software Vendors

We’ve had multi-core chips for a while now, and it’s really just the beginning.  As Gary put it, if we don’t get multi-core chips working Moore’s Law stops - this is not a hardware problem, but a software problem.  Or put another way, a problem shared by both domains, but the software half of the house is behind the curve.

It’s not so much the putting of more gates on the die, but getting more of the gates working more of the time, at the same time.  Oh, and using less power while doing it.  Among many things, this requires a whole new software development tool chain, most crucially compilers and debuggers.

Apparently in discussions with companies like Green Hills, they haven’t come to grips with the problem and don’t really understand the need.  It would be interesting to hear from someone who has dug below the marketing on Green Hills.  Their press releases talk about supporting multi-core processors, but the MULTI IDE (Integrated Development Environment) description makes it look like “MULTI” refers to the plurality of tools integrated into the IDE and not specific support of multi-core processors.

So, the winners in this space could well be in a garage somewhere in Eastern Europe cranking away on a compiler as we speak.

Anyone out there?

Opportunity #2: Integration of Embedded Software

Closely linked to Opportunity #1, this is an area where EDA could really help itself. The most powerful way to communicate value is through your customer’s customer.  If your products address a need or solve a problem for your customer’s customer, you’ve got a winner.  EDA has never really been very successful at this, which may be the main reason why we have seen little progress in EDA’s share of overall electronic design total revenues.

Bottom line: A company that combines both EDA and Embedded software technology and competence will be ideally placed to take advantage of this growing opportunity.  On the face of it, out of the “Big 3″ Mentor is clearly best placed from a product portfolio perspective.  Mentor acquired the Nucleus OS back in 2002 when they bought Accelerated Technology.

Apparently Mentor “gets it”.  Have you seen this investment paying off?

I asked Gary why the industry hasn’t appeared to really engage Wind River Systems. Apparently they were courted a few years ago, but it didn’t go anywhere - apparently they “didn’t get it”.  They have more proximate issues to address, including the remaking of their immediate market following the success of the open source Linux movement and their subsequent re-positioning of the business as Device Software Optimization.

Gary mentioned that QNX “got it”, and they clearly did.  Back in the ’90’s I worked closely with them as a strategic partner for National Semiconductor.  While a small and feisty independent RTOS vendor they had great technology and an understanding of the importance of the value chain and how crucial a robust ecosystem is to success.  QNX were acquired by Harman in 2004, so that’s probably not going to yield much of an opportunity.

Opportunity #3: The Way we do Business

This came down to the fundamentals of why we built the Xuropa Platform.  Not only do we have technical challenges (moving to a new process node, ESL/RTL integration, etc.), but the electronic design industry is facing perhaps its greatest business challenge.  The global distribution of the electronic design industry supply chain and the sustainability of business models.

Long gone are the days when you could spend a day touring Silicon Valley and “pick-up” all the technology you needed for everything along the electronic design supply chain.  Or network at one of the local watering holes to find an opportunity to sell your IP or EDA tool.

Even within the same company, design teams are distributed all over the globe as companies leverage geographic competitive advantage to remain competitive themselves.  Collectively, we need a distributed “watering hole” where we can find people, products and information to enable the electronic supply chain to operate efficiently.  That is the aim of the Xuropa Platform - to bring the electronic design industry together and make it accessible via your web browser.

Need a resource?  Check out the community and Professional Profiles.  Need some information?  Check out the news feeds, company profiles, product directory or post a request to a forum in a company sponsored Online Booth or Online Suite.  Most importantly, need to experiment with a new tool?  Go to a Xuropa Online Lab and try it out.  No download, no installation, no setup - just your browser and you’re in using the tool seeing if it meets your needs.

Finally, clearly we’ve been straining the EDA business model and it’s time for re-assessment and new ideas.  Software-as-a-Service (SaaS) as a  business model for EDA is such an idea and where Gary Smith and I look at things differently perhaps.

Gary’s point was that tool flows need to stay in-house and that they’re “hardened”.  Indeed, that’s the case for now, but I think industry-wide and commercial efforts have put the infrastructure in place for that to not necessarily be the case.  Also, economic necessity will drive greater decoupling and therefore flexibility.  But that’s a whole other post.

That made for quite a lunch and I look forward to our next discussion!

Posted under Xuropa, industry