The Bank of Finland publishes a paper “Kiss me deadly: From Finnish great depression to great recession”, which discusses the causes of the Finnish Great Depression, 1990-1993. The simulations benefited of big compute capabilities, which were enabled by Techila.
In the project, the authors Dr. Adam Gulan, Dr. Markus Haavio, and Dr. Juha Kilponen estimated a structural vector autoregressive VAR model for Finland, where the shocks were identified by using sign restrictions methodology. This methodology is computationally very intensive for large models and with many sign restrictions.
“Without parallel computing it would have been practically impossible to finalize the project,” the project team argues.
In the project, the authors found that the collapse of the overheated financial and banking sectors starting in 1989 was the trigger of the economic crisis in Finland. Foreign shocks, which include the collapse of trade with USSR in 1991, can account for at most about half of the slump, and these shocks occurred only when the economy was already in free fall. Also, the deleveraging and restructuring process of the financial system substantially prolonged the subsequent recovery. In more general, the project improved our understanding on the interaction between macro economy and financial markets — topic which is very widely debated among academics and policymakers.
Techila Technologies and Cargotec‘s MacGregor have investigated how to optimize the cost of ANSYS HPC system ownership with cloud services integrated to the enterprise IT. This report looks at computing clouds as a platform for engineering simulation, and how to manage the software license costs related to ANSYS computing.
“The role of computing is critical in MacGregor’s business. Nearly all marine structural components are custom designed for a specific application. Tendering stages are quick, and producing a price quote and an initial design proposal need support from the computing. The hectic nature of business causes fluctuating workloads to computing. Building readiness to respond to this challenge requires either a HPC solution, which is designed to the average peak workload, or a new kind of HPC solution, which can provide the required elasticity by other means,” says Project Manager Jouni Lehtinen, MacGregor Finland.
This study showed that there are scenarios where ANSYS users can improve the TCO of their HPC systems by integrating cloud services to their enterprise IT. These users include businesses, which have highly variable computing workloads in parallelizable computing problems. This Techila Technologies and MacGregor study includes an analysis of eCADFEM‘s Engineering Software on Demand service with the leading public computing clouds Amazon Web Services, Microsoft Azure, and Google Compute Engine.
“Based on the experience of Techila Technologies, problems related to optimization of a large number of design variables, sensitivity analysis, and scenario simulations are good examples of simulation areas, where large-scale distributed computing and cloud technologies can offer valuable benefits,” says Vice President Tuomas Eerola, Techila Technologies.
When performing tests related to this study, the project team noticed that the usability of a cloud-powered ANSYS computing solution can benefit significantly of a solution which orchestrates the enablement of cloud-based processing. An orchestration solution, such as the Techila HPC middleware, can simplify the IT management and end-user experience which will show as a faster ROI and better productivity.
Come meet the Techila team at the ONS2014 in Norway, Stavanger from August 25-28, 2014. We will be located in Pavillion O 1335.
From seismic imaging to risk management, and from process optimization to forward planning, Techila’s technology enables execution of complex and computationally intensive simulations and calculations and you can get the results in near real-time. Let’s meet at the ONS and discuss how your business can benefit of massive scale computing!
Offshore Northern Seas (ONS) provides a platform for the presentation of the political, economic and technological issues involving the international oil and gas industry, as well as showcasing the latest innovations within the industry.
ONS 2014 is a biennial event and will be the 20th Offshore Northern Seas since 1974.
See you in Stavanger!
All clouds are not similar. If the technical features of a cloud do not align with the needs of business, a solution which looks cost efficient can have a high cost of ownership. Based on customer demand, Techila Technologies publishes a report which benchmarks and analyzes the leading clouds in real-world high-performance computing (HPC) scenarios: Amazon, Google, and Microsoft Azure.
The goal of Techila Technologies’ benchmark experiment was to provide HPC customers with an easy-to-understand analysis and benchmark. Many users who are considering cloud as a HPC solution have said that a FLOPS/ USD and Gbps/ USD figures are interesting but they do not show the full picture. Raw processor power, available memory, or theoretical maximum data transfer rate do not always translate directly to application performance. Because of this, the focus of Techila Technologies’ benchmark experiment is on testing the performance of AWS, Google Compute Engine GCE, and Azure in real-world HPC use-cases, and on studying how the leading clouds can respond to requirements arising from HPC scenarios.
We can say without doubt that cloud computing will have a role to play in the future of HPC. it is also clear that cloud will shape the way, how we will understand the word “HPC” in the future. But cloud will also require learning new ways of thinking. Understanding the needs of business and understanding the features of the cloud will define successful implementations of cloud-powered HPC. IT managers who have this knowledge will be able integrate cloud services in a way, which can serve the fluctuating needs of business efficiently.
Marko Koskinen from Techila Technologies’ Support Team delivers a series of three short tutorial videos on how to use Techila HPC middleware to speed up MATLAB applications without MPI or CUDA programming skills.
In the first part of the series Marko covers how to access the Techila environment from MATLAB. The computing infrastructure of the Techila environment can consist of servers or cluster nodes in your Intranet, computing resources from a cloud, or even in a hybrid system which scales from on-premises server to the cloud.
When we have learned how to enable Techila features in MATLAB to scale out our code to many computers and we know how to parallelize MATLAB applications, Marko will introduce how to use Techila with data programming. In this second tutorial video Marko shows the basics of distributed MATLAB data analysis using the Techila HPC middleware.
In the third part of this tutorial series Marko shows how to debug and do troubleshooting error situations in a distributed computing environment.
A more comprehensive Techila with MATLAB user guide can be found in the technical documentation included in the Techila SDK. Techila SDK can be downloaded from the Techila Extranet. Techila Extranet requires registration, but it is free.
Marko Koskinen from Techila Technologies’ Support Team gives an introduction to Techila Technology. This introduction video covers the very basics of the Techila technology. If you are not familiar with the Techila technology, this technical video provides easy first steps to the solution.
After watching this video you can continue to the video tutorials on Techila’s web site or in Techila’s YouTube channel, or to the technical documentation included in the Techila SDK. Techila SDK can be downloaded from the Techila Extranet. Techila Extranet requires registration, but it is free.
In his presentation, Marko answers questions such as: What is Techila? What is the basic operating principle of the Techila HPC middleware? Why should I consider Techila in my computing, and how do I know if my scenario can benefit of Techila?
CVIS B.V. a provider of ICT services in the Benelux market, today announced that it has entered into an agreement with Techila Technologies. With the Techila HPC middleware technology CIVS can offer its customers a complete and scalable HPC solution for in-house and cloud.
“We are pleased to be extending our relationship with CVIS. We have been actively selling our solutions on the Benelux market for a long time. Now, CVIS will be offering these solutions to its customers, as well. This co-operation with CVIS will now give our customers in the Benelux region the ability to benefit of high-quality service, delivered locally at the local language,” says Tuomas Eerola, Vice President, Techila Technologies.
CVIS provides services for the Research and Development environments in various industries from Manufacturing to Oil & Gas. Due to the market demand and growth of high-performance computing, CVIS wants to extend its HPC service portfolio with a solution, which offers a simple user-experience and deployment in just minutes.
The CVIS solution, powered by the Techila middleware technology, can provide radical speed up to standard and in-house developed simulation and modeling tools, such as MATLAB and ANSYS, and a superior usability and ease of deployment. CVIS will offer the solution at an efficient usage-based model. This will respond to HPC users’ demand to convert HPC costs from capital expenditure to easily controllable operating expenses.
Get ready and save the date for conference New Thinking in Finance at Lloyds & Willis head office in London. It is an EU Conference on HPC in Finance with applications to: product pricing, hedging, trading, risk management and asset liability management.
More information about this can be found on the HPCFinance web site.
Request a live Techila product demo and see, how Techila can give rocket speed to your computing!
Let us walk you through the Techila HPC middleware and show how to make heavy computing easier and faster than ever before. Techila can provide standard and self-built tools many hundreds of times speed up.
Techila’s extremely simple-to-use solution hides all the complexity of Cloud and parallel computing from the end-user, bringing computing acceleration to every desk in the enterprise to speed up business-critical data analysis, simulation, optimization and decision-making applications many hundreds of times.
Click here to Request A Demo!
Marja-Leena Linne and other INCF leaders publish a video where they talk how neuroinformatics and computational neuroinformatics brings hope to people who have been impacted by brain diseases and disorders.
The computing technology, which we at Techila have developed can speed up computational neuroinformatics research dramatically. Researchers can cut the computing time of neuronal signaling models to a revolutionary extend with Techila. For example a computational neuroscience simulation, which would have previously taken seven years can now be completed in just six days, without compromising reliability and accuracy of the results.
The Techila HPC middleware plays an essential role in the development of novel data analysis and Big Data methods and models for large biological neural networks. With Techila, the amount of computational testing of biological hypotheses can be increased due to easy access to powerful computing resources.