Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Processors/DSPs

Multiprocessor nets get middleware

Posted: 16 Nov 2005 ?? ?Print Version ?Bookmark and Share

Keywords:loring wirbel? enea? element? multiprocessor? middleware?

Enea is moving into networking middleware with the introduction of a suite of networking services dubbed Element. The developer of the ose rtos will focus on multiprocessing environments based on either network processors or DSPs.

Tony Pearson, VP of product management at Enea, said Elements high-availability and fault-recovery features mean that the software will not compete with protocol-stack software for L3 and L4 networking offered by many vendors. Instead, Element will take the place of middleware that is usually developed by an OEM.

The software provides services that can run on top of OSE, Linux, a hybrid OS Enea is developing called Orchestra and, eventually, even competing RTOS environments such as VxWorks. The communications layer of Element services is centered on discovery and monitoring tools that expand beyond the type of Simple Network Management Protocol (SNMP) tools common in the networking industry.

Element relies on an in-memory database from Polyhedra, a company Enea acquired, for provisioning services that can be extended to carrier provisioning of bandwidth over a network. And it uses an interprocess communications system, Link, which Enea will productize for other environments outside communications.

The Link communications subsystem allows the network processors native processes to discover other processes and synchronize with them.

The processes monitor one another for availability and use publish/subscribe methods similar to those of XML to provide event notification. The Link interprocess communications even help prevent message buffer overflows, Enea said.

The Link tool logs events, relays them to relevant processes and keeps an archive for use in optimizing the networks. High-availability services monitor and detect faults at both the slot and application level.

A policy engine built into Element takes corrective action in the event of faults, such as implementing failovers from one blade to another or rebooting the system when necessary.

Enea uses a shelf and chassis management service compatible with Service Availability Forum and Advanced Telecom Computing Architecture standards. Monitoring tools record details such as chassis temperature as well as power and CPU utilization. The software can manage alarms for the chassis and mirror services for boot servers in a data center.

In addition to SNMP monitoring, the software provides tools for provisioning a network from the node using the network processor or DSP, including the arbitration of access across multiple users.

Element for OSE is currently available in a price range of $75,000 to $600,000, depending on services. Linux versions are in beta now and will be available in Q1 2006, Enea said.

- Loring Wirbel

EE Times

Article Comments - Multiprocessor nets get middleware
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top