GigaIO and SourceCode Launch First Ever Suitcase-sized AI Supercomputer
May 06 2024 - 9:06AM
Business Wire
Co-designed for portability and scalability,
Gryf is airline cabin baggage-compliant and purpose-built for
go-anywhere insights at the edge.
GigaIO and SourceCode today announced the launch of Gryf, the
industry’s most powerful mobile AI platform for tactical edge
environments, at GEOINT 2024. Billed as “Gryf, co-designed by
GigaIO and SourceCode,” the system is the result of a unique
technology collaboration between the companies that is designed to
engineer complex customer solutions. GigaIO provides award-winning
open workload-defined infrastructure for AI and accelerated
computing, and SourceCode is a global provider of co-designed,
compliant IT systems for next generation intelligent
infrastructure.
Gryf sets a new standard for on-demand configurability in the
field, and is unique in its ability to scale to the performance of
a field supercomputer, all in the TSA-friendly form factor of
carry-on luggage. This scaling is made possible by GigaIO’s
groundbreaking FabreXTM AI memory fabric. FabreX can disaggregate
and re-aggregate GPUs and other system components in order to
enable optimal hardware configurations for each workload. For
example, a single Gryf system can support server-class GPUs to
process over a petabyte of information. This combination of power
and portability led to the Gryf product name, which is derived from
the mythical gryphon that combined the body of a lion with the head
and wings of an eagle.
Gryf allows users to quickly transform the vast amounts of
sensor data collected at the edge into actionable solutions. Prior
to Gryf, data either had to be sent to a datacenter for processing
and analysis, or be limited to the storage capacity that edge
computing devices without GPUs can handle, severely limiting the
ability to produce relevant, on-site results.
Highlights of the unique capabilities Gryf offers include:
- A singular, composable solution for a diverse set of mission
needs. Gryf offers a single, adaptable solution for a variety
of applications, ensuring users can meet changing field
requirements. Fully configurable on-demand through software or by
interchanging sleds, Gryf allows for dynamic reconfiguration,
resulting in higher resource utilization, reduced hardware costs,
higher availability, and lower power consumption.
- Maximum compute density for unparalleled AI performance at
the tactical edge. Gryf sets a new standard for compute density
in a portable format. By overcoming traditional cooling challenges
through disaggregating power-hungry GPUs and providing for its own
ventilation, Gryf is able to combine high-performance GPUs with
substantial storage capacity (up to a petabyte) to allow users to
execute complex AI tasks more quickly and effectively, directly at
the operational site. Each Gryf chassis can accommodate a mix of
six compute, accelerator, storage, or network sleds.
- Easily scale to meet diverse mission demands and boost
operational responsiveness. Up to five Gryf units can be
seamlessly interconnected using GigaIO’s FabreX AI memory fabric
for processing of petabyte-size data sets. This configuration
allows each Gryf unit to access and share sled components across
all five connected chassis in an all-to-all topology.
- Easily transportable for field-missions. Designed for
true mobility, Gryf features a rugged, roll-on TSA-friendly form
factor that fits into an overhead bin to allow for deployment at
any location.
“With AI becoming more pervasive, there is a growing segment of
the edge market clamoring for a meaningful amount of accelerated
compute locally,” said Alan Benjamin, CEO of GigaIO. “This is true
for our Department of Defense customers, who have emphasized the
critical need for timely and actionable intelligence in the field.
Gryf’s novel architecture, made possible by FabreX, our AI memory
fabric, provides those customers the advanced compute, storage, and
GPU capabilities they crave in today’s sensor-rich edge
environments.”
“Engineering this level of power into such a compact form factor
for mission-critical applications is a massive achievement,” said
Parag Shah, CTO/COO of Edge and Embedded AI, SourceCode. “This is
possible due to a deep understanding of customers’ needs, decades
of mission-focused engineering experience, and the collaborative
power of co-design.”
Complex Needs Demand a Co-Design Approach
The co-design approach redefines the solution creation process
by bringing the customer together with technology experts and
best-of-breed products to build and deploy the optimal solution.
The Gryf co-design partnership marries GigaIO’s deep expertise in
interconnects and composable infrastructure solutions with
SourceCode’s ruggedized system design and co-design
manufacturing.
Gryf is available now for purchase from both GigaIO and
SourceCode, with delivery in Q3; early access units configured for
intelligence community workloads are scheduled for June. Gryf is
manufactured in SourceCode’s US-based ITAR compliant co-design
manufacturing facility, part of its global network of manufacturing
sites. Visit SourceCode’s booth (#109) at GEOINT 2024 in Kissimmee,
May 6-9 for more.
About GigaIO
GigaIO provides workload-defined infrastructure through its AI
memory fabric, FabreX, which seamlessly composes rack-scale
resources and integrates natively into industry-standard tools. The
SuperNODE is an “impossible server,” fully engineered to “Just
Work” for AI and accelerated computing. Our solutions allow users
to deploy systems in hours instead of months. Visit www.gigaio.com
or follow X / LinkedIn.
About SourceCode
Founded in 1992, SourceCode is a global provider of co-designed,
custom, certified IT systems for next generation intelligent
infrastructure. The company’s edge devices, desktop, servers,
storage, clusters, and reference architectures are available via
its industry-leading ecommerce platform as well as its
customer-centered engineering-led engagement model. Learn more at
www.SourceCode.com.
View source
version on businesswire.com: https://www.businesswire.com/news/home/20240506933173/en/
Danica Yatko 760-487-8395 danica@xandmarketing.com