In-house Versus In-cloud, That’s The Question
An UberCloud survey on LinkedIn about engineers’ concern about cloud computing revealed that many engineers tend to compare the benefits of workstations versus servers versus clouds in a somewhat misinformed way. In fact, most of them compare the positive aspects of their workstation with some of the (more apparent) roadblocks of the cloud. Example: data transfer; there is obviously no data transfer necessary when you compute your task in your workstation, and yes, there often is heavy data transfer from the cloud back to your workstation (often in the order of several GBytes of result data). But this is simply due to the larger computations in the cloud which are impossible to do on our workstation. And the transfer speed highly depends on our last mile connection. This we can change…

First, we should answer this question: is my workstation big enough and fast enough for the kind of problems I want to solve? If your answer is YES, it is, then you don’t need an HPC Server at all, and you don’t need HPC in the Cloud, full stop.

But if your answer is NO, it isn’t, my workstation is not big enough and fast enough for the kind of problems I want to solve, then a reasonable way to look for viable alternatives is to compare the two options and check which one is most reasonable for you: in-house HPC Server versus remote HPC Cloud, and NOT versus your own workstation which already proved to be useless for your more complex, more challenging tasks. Servers against Clouds! We are certainly fully aware that such a comparison has to come along with some generalizations and simplifications.

Procurement: Bound to approval from upper management; lengthy process: purchase planning, standards determination, specifications development, supplier research and selection, value analysis, financing, price negotiation, making the purchase, supply contract administration, inventory control and stores, waiting in line, accepting delivery, installing and certification testing the hardware, training people, and other related functions. On the other hand, cloud services are short-term on demand or on reservation available for comparatively low cost.

Budget: CAPEX related assets have to be approved often by upper management, while OPEX usually falls into the responsibility of mid-management or even the employee.

Operations, maintenance: To run a compute server, one needs specially trained people; regularly upgrade system and application software; handle and fine-tune the system, workload and resource management; deal with power consumption, cooling, and room temperature; take care of downtime and user productivity. In contrast to cloud where none of these efforts apply.

Flexibility: You bought the system, and you stick with it for three years, even if for some applications your system might not be optimal. Completely different with clouds: there is flexibility in the choice of hardware, software, related tools, timing, pricing, utilization, and so on.

Agility: Users can self-service against a large and flexible service catalog. They can pick up whatever they need, whenever they need it. And, long wait queues can kill engineering inspiration.

Reliability: One single system result in one single point of failure. With clouds reliability can easily be achieved by working with several cloud providers.

Average utilization: Especially in small and medium enterprises, server utilization is unpredictable, because of different project deadlines, the engineers’ vacations or business trips, and weekends where these servers are often almost completely ‘jobless’, resulting on low server utilization. In clouds, in contrary, prices per core per hour are calibrated with high utilization assumed.

Security: Please see our next paragraph about cloud security.

Technology: New technologies and products are coming to market at fast pace. To make up for this we have to regularly upgrade our existing equipment and thus invest even more money. Then we have to stick with our existing systems for at least throughout the depreciation phase. Clouds: to stay competitive providers are regularly refreshing their infrastructure. Therefore, in the cloud, we can shop around for the fastest and best suited hardware and services.

Data transfer: Here we should differentiate between intermediate results and the final dataset: Intermediate data can often be stored in Dropbox or Box.com which have fast connections to the clouds. For final datasets there are transfer technologies which compress and encrypt the data, can send it in parallel chunks, stream it back to the user, or applying data reduction from VCollab.

Full control over your assets: With the advent of the UberCloud containers, additional functionalities like collecting granular usage data, logs, monitoring, alerting, reporting, and more are bringing back the control a user wants.

Software licensing: Due to customer demand and competitive forces, more and more ISVs are now adding flexible cloud-based licensing models, e.g. for monthly, weekly, daily, or even hourly usage.

Access: In the meantime access to many clouds can be considered seamless, and is included in the bill.

Wait time: With your own compute server during peak loads when you need your server the most, your jobs are sitting in the wait queue for hours at end. Clouds change that, simply because Clouds offer “infinite” resources; and if the resources of one cloud provider are not “infinite” enough, you can move on to the next cloud provider. Clouds inherently have very short or no wait time at all.

An Honest Word About Security in the Cloud…
Here we quote a recent article from Mike Kavis, Vice President & Principal Cloud Architect at Cloud Technology Partners [11] about security of SaaS solutions, who refers to a study from Alert Logic [12]. What this report proves is that the security threats are the same, regardless of where the data resides. What is even more interesting is that the success rate of penetrations from outside threats was much higher in enterprise data centers than in external cloud environments. Based on this information, skeptics should dismiss the notion that data cannot be as secure in the cloud as it is behind the corporate firewall. Mike Kavis call this the “hypocrisy” of enterprise IT. It is almost comical when people declare SaaS to be unsecure while their company transmits unencrypted email, staff members have company information on personal unsecured mobile devices, and a number of systems run on unsupported or unpatched versions of software on premises.

… and About HPC Cloud Software Licensing
HPC Cloud software licensing is still considered as one of the major roadblocks for the wider adoption of cloud computing especially for SMEs, as stated by a poll during a recent UberCloud Webinar: the ISVs’ slow adoption of more flexible (on-demand) licensing models for the cloud was the major concern of 61% of respondents.

Beside those software providers who are slow in providing on-demand licenses, for example those who still are providing their keys via dongles, the majority of especially the larger software providers are either working on a professional on-demand licensing strategy, or they already offer Software as a Service. In the first group of those which are preparing software licensing on demand are ISVs like Simulia, and in the second group we find ANSYS with their recently announced Enterprise Cloud, Autodesk with CFD Flex, CD-adapco with Power on Demand tokens, and COMSOL with their cloud-ready network license.