What kind of applications use SIMtone HPC software?

We have some very critical single thread Windows applications, can SIMtone ASPEED help us use multi-core to reduce response time?

How can SIMtone's ASPEED product help us move our large Solaris-based application with a very complex and large shared memory data structure to a Linux cluster?

We are moving to a grid, how can SIMtone HPC products help?

How long does it take to deploy applications using SIMtone's ASPEED? 

We use DataSynapse, Platform or United Devices, why should we consider using SIMtone's ASPEED?

I have heard SIMtone's ASPEED helps more effectively use the capacity provided by dual-core and quad-core platforms, how does ASPEED do this?

Does SIMtone's ASPEED solutions support Java and C#?

We already use a SOA approach to applications, where does SIMtone's ASPEED fit?

We have some very computational intensive applications which we distributed using MPI, how can SIMtone's ASPEED help with these?

We have some very sophisticated deterministic applications  that aren't currently distributed, how can SIMtone's ASPEED help?

We are already distributing our applications using J2EE application servers, what would SIMtone's ASPEED add to this solution?

----------------------------------------------------------------------------

What kind of applications use SIMtone HPC software?

SIMtone's ASPEED products can be used to significantly reduce the run time of high performance and iterative applications in:

Financial Service institutions of all types for applications that need to consider more variables, finer granularity and more precision in a compressed window of time. Financial Services sectors using SIMtone' s ASPEED software today include:

Capital markets for portfolio and credit risk, derivatives, CDOs, trading decision support, data mining, front, middle, and back end office applications.

Retail banking complex applications which execute credit scoring and fraud detection, data mining, risk management, fraud detection, anti-money laundering and marketing analytics. 

Asset Management for significantly reducing the computation times for risk analysis, trading decision support, asset/liability management and portfolio optimization applications. 

Insurance markets for improving the precision of asset liability analysis and actuarial models by supporting a greater number of model points modeling and pricing asset liability, dynamic liability projections, and cash flow/valuations.  

Life Sciences specifically focusing on Pharmacokinetics and Pharmacodynamics efforts in support of late stages of new drug development. For example, SIMtone's ASPEED provides a patented method that  reduces the time of individual Nonlinear Mixed Effect Modeling programs thought to be unparallelizable.

 

Energy / Manufacturing / Government supporting Monte Carlo, Non-recombining Tree and quasi Newtonian, explicit and implicit (direct and iterative) numerical schemes resulting from Finite Element and Finite Difference Methods of approximating the solutions of PDEs, as well as unconstrained (Line Search Methods, Newton Methods, etc.) and constrained (Linear and Quadratic Programming Methods, etc.) optimization problems.

We have some very critical single thread Windows applications, can SIMtone's ASPEED help us use multi-core to reduce response time?

SIMtone's ASPEED software offers a way to parallelize and distribute a single thread application. Depending on the application this could entail some basic set-up without any application changes or some minor annotations to the application. These efforts enable the application to exploit multi-core, multiprocessors,  cluster or grid configurations yielding significant performance improvements and immediate improved robustness.

How can SIMtone's ASPEED product help us move our large Solaris-based application with a very complex and large shared memory data structure to a Linux cluster?

SIMtone's ASPEED software offers a way to quickly adapt and distribute SMP applications without changing the application logic. SIMTone's ASPEED easy to use adapters mask the application from the issues associated with distribution and ensure that the application flow and algorithms are left intact. SIMtone ASPEED accelerated applications yield significant performance improvements, enable the application to exploit multi-core, multiprocessors,  cluster or grid configurations  and immediate improved robustness.

We are moving to a grid, how can SIMtone HPC products help?

SIMtone's ASPEED software can quickly enable your key long running applications to be tested and run across multiple CPUs and or systems prior to implementing the grid infrastructure. This enables you to see the anticipated performance improvements enabled by the grid capacity without affecting the integrity of the original application logic or results. SIMtone's ASPEED software does not require any infrastructure software to be installed.

How long does it take to deploy applications using SIMtone's ASPEED? 

Based on our customers’ experience with implementation, an application can be ready for testing and validation in hours or days depending on the size, complexity and run time characteristics of the application.  The actual deployment is measured in minutes to hours depending on the specific configuration and infrastructure.

We use DataSynapse, Platform or United Devices, why should we consider using SIMtone's ASPEED?

If you have grid deployed applications with inner loops or applications with interdependent data  or complex algorithms (e.g., mesh, PDEs, Finite Element, Finite Differences, non-linear regression) that are challenges to deploying on the grid, SIMtone's ASPEED can provide an immediate solution to achieving significant predictable performance improvements while more efficiently exploiting the grid.

I have heard SIMtone's ASPEED helps more effectively use the capacity provided by dual-core and quad-core platforms, how does SIMtone's ASPEED do this?

SIMtone's ASPEED software can be readily used to parallelize the entire application or specific application components essentially enabling it to run multiple parallel threads. We have seen cases where VBA applications have run 4 times faster on dual core processors with hyperthread on. This can be readily achieved without the need to reprogram the application to use multi-threading.

Does SIMtone's ASPEED solutions support Java and C#??

Yes, the SIMtone ASPEED SDK includes both Java and C# Adapters that can be easily inserted in existing or new applications.

We already use a SOA approach to applications, where does SIMtone's ASPEED fit?

SIMtone's ASPEED is simply a SOA performance expansion. It enables you to easily run applications with parallelized components as part of the application architecture. For example an Excel program can actually be doing its parallelized calculations on a Linux cluster or grid. The same calculations can also be used by runs submitted as batch runs directly to the Linux cluster.

We have some very computational intensive applications which we distributed using MPI, how can SIMtone's ASPEED help with these?

SIMtone's ASPEED adapters are similar to MPI in that they enable the communication and control among distributed programs, however that is where the similarity ends. SIMtone's ASPEED actually provides a much higher level interface which automatically takes care of the launching, management and distribution of the ranges and data under the covers. This enables the programs to react more dynamically to configuration and real-time load changes without having to focus on the distribution mechanics. Since the application is already parallelized, it is very easy to insert the SIMtone ASPEED adapters and comment out the MPI.  The resulting application will be able to be distributed across work groups, clusters and grids without the need to have any daemons or MPI already deployed to the systems and significantly enhance performance.

We have some very sophisticated deterministic applications  that aren't currently distributed, how can SIMtone's ASPEED help?

SIMtone's ASPEED provides adapters that support non-linear , tree and other stochastic and deterministic applications. The adapters mange the data partitioning, range distribution, data movement and synchronization automatically.

We are already distributing our applications using J2EE application servers, what would SIMtone's ASPEED add to this solution

Using the SIMtone ASPEED Java adapter, applications can dynamically spawn the iterative methods across the current system and other systems adding a new and important performance dimension to a J2EE application.