Abaqus is one of the leading finite element analysis (FEA) solutions in the industry. Many of us will have used it either on a local workstation or with the aid of high-performance computing (HPC) on a local network. But what if you don’t have time or budget to research and setup the perfect configuration for your simulation needs? Well, fortunately, Dassault offers the option of computing in the cloud. That means no hardware cost, no maintenance, no need to periodically upgrade. It can also be used as a burst resource, meaning that you’re not fighting with your colleagues for tokens and that very large explicit job (that might take days on your local machine) can be run in minutes or hours. Let’s dive in and take a look at exactly how it works, and what you need to do to get started!
Consistent vs Consumable Models
On-cloud consistent token licensing works in much the same way that our traditional Abaqus tokens work. You purchase a certain number of tokens and they allow you run on the SIMULIA cloud at no additional charge. Having access to more tokens permits running on more cores or running more job simultaneously.
Consumable credits are exactly that – consumable. You will purchase a block of them and they are ‘spent’ as the analysis runs. The more cores you choose to utilize, the greater the credit cost per hour of runtime.
The table below shows typical numbers of cores and their associated cost in terms of both consistent cloud tokens and consumable cloud credits:
When To Use Consistent Tokens
Consistent tokens work best when you’re part of an organization that’s going to be running analyses very consistently, day in, day out. Obviously, any down time represents sunk cost that cannot be recovered.
When To Use Consumable Credits
On the other hand, consumable credits are a great option if you’re workload is sporadic, or if you’re just not doing that much simulation (yet). With these, you’re operating on a pay-as-you-go basis, so if there are long periods of time between simulations, you won’t be paying for the privilege. Of course, there is a break even between these two models, and that’s something that a Dassault partner like Fidelis can help you understand during the purchasing process.
It should also be noted that consumable credits be very valuable when you want to run a large number of analyses, say for a DOE. With traditional tokens, you’d need to stack these sequentially, which means the wallclock time to complete is significant. On the cloud, however, you can run analyses simultaneously at the same cost to you. This can result in really significant time savings as can be seen in the plot below. Here, we compare the total wallclock time needed to run a battery of 2 hour simulations sequentially (using traditional token-based licensing) and simultaneously (using cloud credits).
When To Use A Hybrid Model
A hybrid of the two aforementioned licensing models can be particularly valuable when your organizational workload is volatile. For example, imagine that you typically have one engineer working on simulation, but you hit a period of peak demand and need two or three engineers to be running jobs simultaneously – your four (eight? sixteen?) core install might not be enough to keep up. That’s where having a moderate number of consistent tokens and a backup stash of consumable credits might come in useful. The plot below illustrates where having cloud credits on deck can be really handy.
Final Thoughts
The world is changing – and it’s time that the way we think about simulation software does, too. No longer do we need expensive hardware and perfectly configured installs to be able to keep up with the demanding nature of the job at hand. On-cloud simulation solutions provide a more dynamic and versatile option that can save both time and money for businesses just like yours!
If you’re interested in learning more about Abaqus on the cloud, don’t hesitate to reach out to our expert team at Fidelis – we’ll be sure to point you in the right direction!