As I mentioned, we had a record quarter for CRAY J90 Series system
sales in the three months ending September 30. These systems remain the
fastest selling product in Cray's history, with some 350 sold to date.
In other product lines, the news is equally encouraging. Since its
introduction, we have shipped 35 CRAY T90 Series systems. Of the total
shipments, 11 are large-configuration CRAY T916 or T932 systems. We are
pleased to see this balance in the CRAY T90 family of products.
Sales of CRAY T3E Series scalable parallel systems have significantly
exceeded our own expectations. By the end of October we had shipped some
30 systems to eight countries. One-third of the systems shipped to date
have been high-end configurations with 128 processors or more.
O n November
11, we introduced the CRAY T3E-900 Series, the successor to the CRAY T3E
featuring faster processors, improved price/performance, and
record-shattering performance that makes this system the first and only
commercially available teraflops supercomputer on the market. Orders for
CRAY T3E and T3E-900 systems now exceed $250 million, making this product
by far the best selling scalable parallel supercomputer.
Customer interest in all our product lines remains strong, as evident
in orders over the past six months from customers such as Boeing, Honda,
the Kyoto University Institute for Chemical Research, the United Kingdom
Meteorological Office, Mitsubishi Electric Company, several Department of
Defense sites, and NASA/Goddard, among others. In addition, during the
quarter we learned that, together with Silicon Graphics, we had been
awarded the single largest procurement to date under the U.S. Department of
Energy Accelerated Strategic Computing Initiative (ASCI).
We believe the Department of Energy's award of this $110 million
contract to Silicon Graphics and Cray Research recognizes the unique
capabilities the combined companies bring to the market for
high-performance scientific and technical computing. On this particular
procurement, it is abundantly clear that the combined Silicon Graphics/Cray
Research solution was superior to what either company could have offered
independently.
When Silicon Graphics and Cray Research closed on the merger of the
two companies in 1996, we noted how the combined company is
uniquely well-equipped to deliver a "desktops to teraflops" family of
products. In October, we started on our journey toward that goal with the
introduction of the Origin Series of systems based on a new architecture,
scalable shared-memory multiprocessing (S2MP).
The Cray Research piece of this new product family is the CRAY
Origin2000 Series of supercomputers. The CRAY Origin2000 Series is an
addition to the overall Silicon Graphics offering for high-end customer
requirements and complements existing Cray Research products. These
systems will provide customers with unmatched scalable performance on
third-party technical, commercial, and visualization applications in a
platform configured with from 65 to 128 MIPS processors.
N ear-term, there is little overlap in the
market and applications for
CRAY Origin2000 Series systems and the markets and applications for our
existing CRAY T90, CRAY J90, and CRAY T3E Series systems. The CRAY T90
and J90 products remain the best platforms for a broad range of vectorized
third-party applications, while the CRAY T3E-900 Series provides the extreme
scalability required primarily for proprietary, customer-developed
applications optimized for highly scalable platforms.
Longer-term, however, we expect the S2MP architecture will become the
basis for a unified, scalable, high-performance platform. This unified
architecture will integrate leading-edge simulation, visualization, and
data-management capabilities to provide a means of solving the world's most
demanding scientific and engineering problems.
As has been the practice at Cray Research and Silicon Graphics, we
will collaborate with you, our customers, to achieve the breakthroughs
required to meet your emerging needs. The merger with Silicon Graphics
enables us at Cray to focus completely and intently on high-end
supercomputing while the combined strengths of Cray Research and Silicon
Graphics provide the critical mass of expertise required to deliver an
ideal high-performance platform--one that will enable customers to
achieve new levels of scientific, engineering, and business insight through
interactive, integrated simulation and visualization.
In consultations with you, here's how we have heard the attributes of
an "ideal" platform defined:
- Near-linear scalability offering low entry prices and "pay as you go"
pricing, as well as the ability to achieve extreme levels of performance
- A "general purpose" design offering broad application compatibility
- A shared-memory architecture offering ease of programming
- Integrated visualization and data management resulting in intuitive
system use
- Cray-class single-processor performance, resulting in sustained Gflops
performance
A unified architecture addressing these
requirements will be
introduced in phases over the next several years, but the journey began
with the launch of the S2MP architecture last fall in the form of the
Origin Series.
The next-generation S2MP product (code-named "scalable node
one" or SN1) will provide scalability to thousands of processors as the
extreme scalability now available only in CRAY T3E systems is incorporated
into SN1 in the form of "globally scalable" interconnect, memory I/O, and
operating system capabilities.
In addition, SN1 will achieve a performance
improvement resulting from a step-up in the speed of the binary-compatible
individual processors. The SN1 implementation of the unified architecture
is expected to be available before the turn of the century.
The third generation S2MP product (SN2, targeted for availability after
2000) brings together the key attributes desired in an ideal platform in a
single architecture. SN2 will be the first highly scalable system that
combines the price/performance benefits of RISC processors with the
single-processor performance of our leading vector processors.
In meeting this challenge, the combined strengths of Cray Research and
Silicon Graphics will prove especially valuable. Together, Cray and
Silicon Graphics bring to this task distinctive expertise in uniprocessor
design, including vector processing, and a microprocessor design team
dedicated to high-performance solutions.
In short, our combined company
has the resources and, equally important, strong business and economic
incentives, to develop the "Cray on a chip" processor that will be the
critical enabling technology for our ideal platform. Offering tightly
integrated computation, visualization, and data-management capabilities, SN2
will be the successor to all current and planned Cray Research and Silicon
Graphics high-end product families.
A s high-performance hardware
architectures migrate to a unified
platform, so too will the software environment, which will converge on a
common, scalable operating system architecture. Working together, Cray
Research and Silicon Graphics intend to meld complementary capabilities
from UNICOS and IRIX to create a "best of both worlds" environment that
will protect customers' software investments, mitigate migration issues,
ensure availability of a broad range of applications, and provide an easy
to-use programming model.
As the evolution to a unified high-performance architecture unfolds,
Cray Research and Silicon Graphics will provide you with multiple migration
paths. Our recent introduction of the CRAY T3E-900 Series, hand-in-hand
with the commitment to further microprocessor upgrades, is representative of
how the transition will be handled. CRAY T3E-900 Series systems will merge
with the next generation of CRAY Origin2000 supercomputers in the 1998-1999
timeframe as SN1 becomes a fully supported product.
Nevertheless, we are
planning another enhancement to CRAY T3E performance beyond that just
introduced to provide an upgrade path for customers who do not plan to
replace their CRAY T3Es until after 1998. Similarly, CRAY J90 Series
customers will be able to take advantage of planned upgrades to the series
or migrate to SN1 or SN2 implementations of the unified architecture over
the next several years.
In terms of single-processor performance, CRAY T90
Series systems will remain the fastest systems offered by Cray Research and
Silicon Graphics through the end of the decade. Work continues on a
follow-on product to the CRAY T90, whose logical successor will be the SN2
product, which becomes available some time after the year 2000.
A combined service organization will support this unified product
family. Recognizing Cray's leadership in providing outstanding customer
service, Silicon Graphics chose Cray's Mick Dungworth to head the service
organization for the entire company. Mick and his team are already at work
on creating a combined service strategy that recognizes the very large
percentage of you who are both Cray and Silicon Graphics customers.
In our journey toward a unified architecture,
CUG will continue to be
an important sounding board and source of ideas and inspiration. After all,
it is your collective aspirations to advance your respective sciences and
achieve new insights to benefit your organizations that drives us to
deliver ever higher levels of computational power.
The combined vision of
Silicon Graphics and Cray Research is to meld simulation, visualization, and
data management to enable you to solve problems others can't even imagine.
We're convinced no company is better positioned to make that vision
reality, and we're eager to prove that to you through the breakthroughs
we'll be bringing to you in the years ahead.
