Gemstone Systems Launches GemFire Enterprise 5.0
Next generation data fabric delivers massive scalability and improved performance for distributed systems
Beaverton, OR -- September 12, 2006 -- GemStone Systems today introduced GemFire 5.0. Designed specifically to meet the demanding performance and scalability requirements of advanced J2EE, grid, SOA, and OLTP applications, GemFire Enterprise 5.0 provides a high-performance, distributed operational data store that enables real-time operations across distributed locations. The scalability features of GemFire Enterprise 5.0 enable organizations to manage massive data loads while guaranteeing data integrity and availability to power mission-critical applications. By enabling applications to scale transparently, GemFire Enterprise 5.0 guarantees operational reliability while preventing single points of failure and managing exceptions without widespread impact.
“Enterprises are facing the growing challenges of using disparate sources of data managed by different applications,” noted Noel Yuhanna, Forrester Research. “New technology is emerging that Forrester has coined ‘information fabric,’ a term defined as a virtualized data layer that integrates heterogeneous data and content repositories in real time. The potential benefits of this technology are so great that enterprises should develop a strategy to leverage information fabric technology as it becomes more widely available.”
As the core component of the GemFire Enterprise Data Fabric, this release combines essential components of databases, messaging and caching, providing distributed data virtualization and sharing to counter the real-world deployment challenges of complex application systems, wherein a single offending member can bring the entire system to a grinding halt. GemFire Enterprise 5.0 provides scalability along three distinct elements - data scalability, distribution scalability, and cluster/membership scalability -- without compromising performance.
“SOA and grid computing projects frequently present dramatic data management challenges, requiring reliable access and delivery of data among disparate applications with minimum latency,” said Shankar Iyer, executive vice president, marketing and strategy, GemStone Systems. “The new features and improved functionality of GemFire 5.0 reliably and consistently manage data persistence and distribution across several hundreds or thousands of nodes, providing a distributed data layer to maximize deployments.”
Leveraging standards based querying and optimized, native data management, GemFire Enterprise is built on one of the industry’s fastest and most reliable data distribution and caching systems. New features of 5.0 include:
Role-based reliable data distribution: GemFire Enterprise 5.0 boasts innovative role-based reliable data distribution, providing a user-defined method for managing data routing while offering advanced levels of reliability and consistency across large scale, dispersed deployments. This feature enables application architects to define “roles” related to specific cache members and further identify certain roles as “required roles” for a particular cache node. In this manner application dependencies can be modeled at the infrastructure level to guarantee reliability and data consistency in a configurable manner.
Enhanced data portioning model: By combining previously available data portioning schemes with data region semantics used to define and describe logical data namespaces, 5.0 enables interoperability between partitioned and non-partitioned regions within a distributed system. User-defined policies/configurations control the memory management and redundancy performance of the partitioned regions to handle node failures, redirecting the affected client requests automatically to a back-up node, allowing architects to dynamically scale their applications to increasing data loads.
Loosely coupled distributed systems: Version 5.0 offers an innovative gateway hub/gateway model to connect and configure distributed systems and/or sites across Wide Area Network (WAN) in a loosely coupled fashion, delivering unlimited scalability. This model enables each GemFire distributed system to assign a process as its gateway hub, which includes multiple gateways that connect to other distributed systems. Back-up gateways and gateway hubs can be created and configured to manage automatic fail-over, ensuring data consistency with no performance degradation.
Advanced management for slow and unresponsive applications: GemFire Enterprise 5.0 provides advanced management for slow and unresponsive applications to maintain a superior quality of service across distributed systems. There are no adverse effects to overall system performance and throughput if any one application/server consumes messages at a slower rate than other receivers. 5.0 allows a distribution timeout to be associated with each consumer, automatically switching from the default synchronous communication mode to an asynchronous mode if message acknowledgements are not received within the timeout period.
Benchmarking Data
GemFire Enterprise 5.0 has been benchmarked in a cluster comprised of 2 CPU, hyper-threaded, 3.6GHz Intel Xeon processors with 8GB of RAM on a 1 Gigabit network. GemFire delivered superior results, including:
Data access rates of 350-400K messages/second - (10K payload)
Data write throughput of 30 GB* in 4 minutes, with 33 server nodes (partitioned caches) populated by 100 clients on a 1GB network
Data distribution throughput of 6000 messages/second to 50 clients, with 4 publishers and a 30 GB data set; in both cases, throughout was capped by network saturation
Multiple partitioned networks, a 10GigE backplane, or Infiniband network would deliver substantially higher throughput.)
Additional enhanced functionality and benefits of GemFire 5.0 include:
Sustainability under network failures/segmentation to automatically manage and recover from such issues
Data region level transport protocol flexibility offering users a choice of transport layer
Client server cache topology enhancements to eliminate cache bottlenecks due to client-side performance and ensure cache scalability
Improved membership model to support linear scalability for hundreds or thousands of cache nodes
Enhancements to cache event model to enable multiple cache listeners to act upon cache events.
Product Availability
GemFire Enterprise 5.0 will be generally available in October, 2006. Please visit the GemFire Download Center at http://www.gemstone.com/download/ to download a preview version of GemFire Enterprise 5.0.
About GemFire Enterprise
GemFire Enterprise provides a pervasive, distributed data fabric for highly available data-on-demand at memory speeds. The GemFire enterprise data fabric (EDF) operates seamlessly on multiple platforms in heterogeneous operating environments with universal language support, enabling data virtualization and transparency of data sources to the consuming applications. GemFire connects to databases, application stores, analytical tools, messaging systems, and mainframes, enabling the deployment of high-performance service-oriented architectures at significantly lower costs.
About GemStone Systems, Inc.
GemStone Systems provides data services solutions for enterprise business architects and data infrastructure managers that are building, enhancing or simplifying access, distribution, integration, and management of information within and across the enterprise.