Increasing demand for applications such as machine learning, artificial intelligence (AI), and video transcoding which are all computationally intensive, has created the need for more powerful yet energy efficient solutions in data centers that can handle large amounts of data cost-effectively. This is where hardware accelerators can play their profound role in efficient processing of target applications. With this new solution the era of heterogeneous distributed computing systems that consist of contemporary general-purpose processors (CPUs), general-purpose graphic processing units (GP-GPUs), and field-programmable gate arrays (FPGAs) has been emerged. Moreover, the global estimated data center accelerator market size of USD 65.3 billion by 2026 shows the importance of accelerators in today’s computing paradigms.
Considering this need, Avan introduced its first service based on its hardware accelerator (code-named Tivan) for video transcoding (https://avanco.tech/services/). The service uses FPGA cards specifically developed for video transcoding alongside servers. The purpose of this service is to free the companies from time-consuming development and reducing their operational costs to maximize their business value through the encoding acceleration of Full HD videos (up to 20x). In addition to these initial goals, the simple and hassle-free use of the service was also considered of high priority so that users can integrate the service into their stack without any worries.
Given the diversity of applications and the needs of researchers, engineers and developers, Avan considered the development of another service in which FPGAs themselves are provided to users for R&D works on the cloud (https://avanco.tech/FaaS/). In this service, users can experience development on a desired device without having to pay a lot of money on FPGA purchasing.
Avan R&D team is currently working on other sections of acceleration such as application deployment, a service that make it possible for companies and start-ups to deploy their own application on the FPGA and take advantage of its benefits.