Accelerating Cloud Applications
with Intel® Xeon® Scalable
Processors and Intel® FPGAs
Cloud Service Providers (CSPs) accelerate data-intensive applications and
workloads for high throughput, excellent power efficiency, and fast response
times using Intel® Xeon® Scalable processors and Intel® FPGAs
Industry Strategic Challenges
The number of connected devices and the ...amount of data that they produce
continues to increase. According to Gartner, there will be 20.4 billion connected
things worldwide by 2020 . The rate at which data arrives from these devices into
data centers also continues to grow.
At the same time, Cloud Service Providers (CSPs) face constraints in their data center
space, power, and cooling and must work within tight margins. Any solutions CSPs
deploy must be power efficient, and support low latency so they can keep up with
the surge in big data. Additionally, they must be programmable and versatile: fixed
silicon accelerators are unable to match the variety of workloads that CSPs face.
By alternating Intel® FPGAs, a wide range of workloads can be accelerated to
accommodate this increased data, and new demands for analyzing it, within the
constraints of the CSP’s data center.
Accelerating Applications with FPGAs
FPGAs are silicon devices that can be dynamically reprogrammed with a data path that
exactly matches your workloads, such as data analytics, image inference, encryption
and compression. This versatility enables you to provision fast processing power
efficiency, and low-latency service – potentially lowering your total cost of ownership,
and maximizing compute capacity within the power, space and cooling constraints of
your data centers.
Traditionally, FPGAs require deep domain expertise to program, but the Intel®
Acceleration Stack for Intel® Xeon® CPU with FPGAs simplifies the development flow
and enables rapid deployment across the data center. Intel is collaborating (partner is
a forbidden word at Intel) with FPGA developers, server OEMs, virtualization platform
providers, operating system vendors and system integrators to enable customers to
efficiently develop and operationalize their infrastructure.
Business Brief | Accelerating Cloud Applications with Intel® Xeon® Scalable Processors and Intel® FPGAs 2
• Ease of deployment. The Intel® Programmable
Acceleration Card (Intel® PAC) provides an Intel® FPGA
in a PCIe-based card. Validated servers incorporating
the card are available from several leading OEMs.
• Standardization. The Intel® Acceleration Stack for
Intel® Xeon® CPU with FPGAs defines standardized
interfaces that FPGA developers and development and
operations teams can use to hot-swap accelerators
and enable application portability.
• Programmability. The Acceleration Stack abstracts
away much of the complexity of programming FPGAs,
making it easier for Tier 1 Cloud Service Providers
(CSPs) and large cloud-based businesses to develop
their own accelerators.
• Off-the-shelf availability. Accelize’s AccelStore*
offers a library of predefined accelerators that next-
wave CSPs can reuse, with charge-back mechanisms
built-in, making it easier to create an on-demand
acceleration service. These accelerators are developed
by an ecosystem of collaborators with deep workload
processing expertise and validated for use by Accelize.
The Intel® Acceleration Stack for FPGAs is a new architecture
that allows software developers to use the
Read the full Business Brief