Press Release

Two New Additions to the Vega 2 Family of Compute Appliances Released

Mountain View , Calif. – June 14, 2007– Azul Systems, the award-winning innovator behind server appliances that deliver compute and memory resources as a shared network service, today announced the availability of two new Vega 2™ Compute Appliances. The 7240 and 7280 models are the most powerful SMP systems in a compact form factor, offering up to an unprecedented 768 processing cores on 16 processor chips with 768 GB of memory. The new Compute Appliances provide the fastest path to maximizing the ROI of business-critical Java applications, by allowing applications to transparently achieve 5X to 50X greater scalability and throughput while delivering consistent performance even under unpredictable load. The new systems further enhance Azul’s product line by offering high-end single-box solutions for challenging business applications in capital markets, telecom B2B and B2C enterprises.

“The 7240 and 7280 pack enough punch to provide breakthrough scalability and performance to handle the demands of the largest business-critical applications,” said Scott Sellers, chief operation officer and co-founder of Azul Systems. “The Vega 2 processor is the only solution with a Java-optimized instruction set, hardware-assisted garbage collection, and support for optimistic thread concurrency – making the Vega Compute Appliances the industry’s first and only platform to deliver the speed and consistency IT departments need to use Java for business-critical and demanding applications.”

“SAP had a need to deliver computational scalability and performance for the analytics application that deals with memory heaps of hundreds of gigabytes,” said Igor Tsyganskiy, head of the Advanced Technology Group at SAP Labs in Palo Alto. “ The Azul Systems Vega 2 7200 system delivered the required throughput and response time due to its scalable memory architecture and advanced garbage collection technology; this type of capability could not be achieved with traditional server technologies.”

According to IDC, sales of enterprise servers to run Java-based workloads exceeded $11.2 billion in 2005 and are expected to continue to grow rapidly, fueled by the explosive growth in Java programmers and the use of Java in enterprise applications. However, business-critical, transaction-intensive applications such as those found at financial institutions, telecommunications organizations, and e-commerce sites require highly scalable compute resources and management capabilities, which alternative server platforms cannot efficiently address. Azul will continue to provide the processing and memory resources required to enable the continuing evolution of large enterprise applications.

“Azul demonstrates unique server leadership for Enterprise Java applications and its capabilities should be reviewed closely," said Cal Braunstein, Chairman and CEO, Robert Frances Group. “The company is taking a fresh approach to overcome the most challenging issues of Java business applications.”

The new Compute Appliances are uniquely positioned to address some of the highest-priority initiatives for Java-based workloads. They are ideally suited to support the demands of Extreme Transaction Processing applications, serve as a consolidation platform for SOA deployments, and power enterprise data grids. In performance tests conducted by Azul Systems¹ ² ³, the next-generation systems, with up to an unprecedented 768 processor cores and 768 GB of memory in a single system, delivered the equivalent performance of 18 to 33 dual-chip, dual-core x86 servers, and 26 to 45 T2000 servers, demonstrating their unique ability to address the demanding needs of enterprise business-critical Java applications.

Availability
The 7200 series compute appliances are available immediately in the Americas. They will be available in Europe in September 2007.

¹ Trade6 benchmark results: Azul Systems Compute Appliance 7280 (16 Vega 2 processors, 768 cores) 23,989 TPS; Sun T2000 (8 cores @ 1.2 GHz) 924 TPS; 2-socket dual core AMD Opteron 285 (4 cores @ 2.6 GHz) 1095 TPS

² XMLMark1.2 benchmark results based on streaming parser workload (SAX1, SAX2 and SAX3): Azul Systems Compute Appliance 7280 (16 Vega 2 processors, 768 cores) SAX1: 20,548 TPS, SAX2: 10,670 TPS, SAX3: 5,470 TPS (average SAX: 12,229 TPS); Sun T2000 (8 cores @ 1.2 GHz) SAX1: 466 TPS, SAX2: 228 TPS, SAX3: 117 TPS (average SAX: 270 TPS); 2-socket dual core AMD Opteron 285 (4 cores @ 2.6 GHz) SAX1: 690 TPS, SAX2: 272 TPS, SAX3: 145 TPS (average SAX: 369 TPS)

³ Enterprise Service Bus results as measured by Azul Systems using BEA AquaLogic Service Bus: Azul Systems Compute Appliance 7280 (16 Vega 2 processors, 768 cores) 24,300 TPS; Sun T2000 (8 cores @ 1.2 GHz) 1,839 TPS; 2-socket dual core AMD Opteron 285 (4 cores @ 2.6 GHz) 1392 TPS

See related releases for more details on the new Vega 2-powered family of compute appliances:

• Azul Systems Next-Generation Compute Appliances Deliver Business Results with Breakthrough Performance for Business Critical Java Applications: http://www.azulsystems.com/press/120406_vega2.htm

About Azul Systems
Azul Systems is a global provider of enterprise server appliances that deliver compute and memory resources as a shared network service to transaction-intensive applications, such as those built on the Java™ platform. Azul Compute Appliances enable transparent, massively scalable infrastructure to support the business priorities of today’s most demanding enterprise environments and deliver increased capabilities, capacity, and utilization at a fraction of the cost of traditional computing models. More information on Azul Systems can be found at www.azulsystems.com.

Legal Notices
Azul Systems and Azul are registered trademarks, and the Azul arch logo is a trademark of Azul Systems, Inc. in the United States and other countries. All Java-based trademarks and their logos are trademarks or registered trademarks of Sun Microsystems, Inc. in the United States and other countries. Other marks are the property of their respective owners and are used here only for identification purposes.

###