Advantech (TWSE: 2395.TW), a global manufacturer of network computing blades, and multicore processor platforms, announced it is now shipping ATCA blades and network appliances employing the Intel® Xeon® processor E5-2600 family, formerly codenamed Romley-EP, to leading network equipment providers and telecom equipment manufacturers for deployment in their next generation platforms. The new platforms feature increased performance, faster and adaptable I/O along with greater memory capacity - all requirements at the core of applications using Deep Packet Inspection (DPI) technology such as policy enforcement, quality-of-service, mobile and video content optimization and subscriber analytics.
Advantech's MIC-5332 dual processor ATCA blade, one of several Advantech products announced today comes with full support for the Intel® Data Plane Development Kit (Intel® DPDK) and Intel® QuickAssist Technology to provide a seamless upgrade path for customer's deploying the company's previous two generations of Intel® Xeon® processor-based ATCA blades. The blade also incorporates a Fabric Mezzanine Module (FMM), a technology at the core of Advantech's Customized COTS framework offering distinct differentiation advantages over COTS platforms and in-house designs.
Advantech also announced their FWA-6510 Network Appliance Series using two Intel® Xeon® processors E5-2600 with 16 physical cores and offering outstanding I/O density scaling up to sixty-four gigabit Ethernet ports or sixteen 10GbE ports in a modular architecture. Network Module offerings allow many port configurations with or without network bypass, offload and acceleration.
"A key aspect of workload consolidation on a single architecture is software reuse. With Advantech's scalable hardware platform ranging from appliances to high end ATCA systems customers have the foundation for greater efficiencies, lower time-to-market and reduced life cycle management costs" said Peter Marek. Director of x86 solutions at Advantech's Networks and Communications Group. "With a single-architecture running control and data plane workloads, the re-use of hardware and software components makes it easy to apply them to new network elements or different Intel processors. It also facilitates the implementation of adaptable and software configurable network elements."
"Faced with increasing CapEx and OpEx, service providers are looking to telecom and networking OEMs to deliver optimized cost/performance ratios and greater energy efficiencies" said Steve Price, Director of Marketing at Intel's Intelligent Systems Group. "Now they have a truly compelling architecture that offers scalable platform choices for consolidating workloads on Intel® architecture, which dramatically reduces development effort, power consumption and time-to-market."
MIC-5322 AdvancedTCA Blade
MIC-5332 ATCA blade enables the highest performance available in ATCA form factor with up to 16 cores and 32 threads of processing power, fast PCI Express gen.3 lanes running at up to 8Gbps, and best in class virtualization support. Two QPI interfaces between the CPUs improve memory and I/O access throughput and latencies when one processor needs to access resources hosted by the other socket. With four DDR3 DIMMs per socket in a quad channel design running up to 1600MT/s, the MIC-5332 not only offers superior memory bandwidth over 3-channel designs, but can also support memory densities up 256GB using latest LR DIMM technology. It outperforms previous generation dual socket designs while keeping similar thermal characteristics with balanced airflow resistance. Using the Intel® C600 series chipset with its integrated 4-port SAS controller, the need for an external storage controller is eliminated making the MIC-5332 an ideal choice for cost sensitive control plane applications.
The blade also introduces the company's first use of Fabric Mezzanine Module (FMM) technology providing expansion capabilities for front I/O or additional fabric ports at significantly lower cost and real-estate than AdvancedMCs. While the blade supports two 10GBaseKX4 interfaces in the base model, dual dual-star fabric implementations can be added by installing an FMM-5001 module. FMM type II sockets with PCIe x16 connectivity provides extension possibilities for additional front port I/O, offload and acceleration controllers such as Intel® QuickAssist(TM) accelerators, IPSec offload engines or customer specific logic. Moreover, FMMs can be reused on RTMs and across different blade designs. This unmatched flexibility combined with the highest performance Intel Xeon processors available make the MIC-5332 equally well suited for both application and data plane workloads.
FWA-6510 2U Rack mount Appliance
The FWA-6510 is a compact 2U rack mount form factor with redundant AC or DC power supplies, up to four 2.5" drives, management I/O and a LCD module. With multiple PCIe x8 pipes balanced between CPU sockets, the system supports up to 4 slots of I/O modules in Advantech's Network Mezzanine Card (NMC) form factor with an 8-slot version in development supporting as much as 160Gbps of IO traffic.
Deep Packet One Evaluation Platform
Based on the FWA-6510 160Gbps-capable 2U appliance, the Deep Packet One Evaluation Platform comes pre-loaded with the Intel® DPDK, the 6WINDGate(TM) software and the Qosmos ixEngine(TM) . The platform enables fast-track evaluation of the latest multi-core technology from Intel and the Intel® Embedded Alliance will allow customers to rapidly develop and test content driven network intelligence applications based on DPI technology. Deep Packet One was introduced to facilitate the technology evaluation process for OEMs by offering a turnkey evaluation platform based on the major building blocks required to evaluate DPI on Intel® architecture. Supported by Advantech's broad hardware offerings, OEMs can begin their development and evaluation work on this turnkey appliance and later move to other target platforms such as ATCA easily.
Charlie Ashton, VP Marketing, 6WIND
"Networking software from 6WIND delivers advanced packet processing functions that are critical for maximizing the performance of DPI-based products on Intel® architecture. With the Deep Packet One Evaluation Platform, developers will discover how multi-core packet processing with 6WINDGate and Intel® DPDK provide performance which scales linearly with the number of cores until the maximum bandwidth of the hardware platform is reached. Processor cores not used to run 6WINDGate are available to run value-added application software or Virtual Machines (VMs), resulting in a highly efficient and flexible system for advanced networking equipment."
Erik Larsson, VP Marketing, Qosmos
"Application developers can accelerate time-to-market by leveraging Qosmos ready-to-use DPI software on the latest Intel® Xeon® Processors. Advantech's Deep Packet One is an ideal platform for enabling the classification of traffic at core network speeds and extraction of metadata in real-time".
The MIC-5332, FWA-6510 and Deep Packet One Platforms are released products and available to order. For fully detailed product and ordering information, please visit the dedicated web page at www.advantech.com/...
Advantech's MIC-5332 dual processor ATCA blade, one of several Advantech products announced today comes with full support for the Intel® Data Plane Development Kit (Intel® DPDK) and Intel® QuickAssist Technology to provide a seamless upgrade path for customer's deploying the company's previous two generations of Intel® Xeon® processor-based ATCA blades. The blade also incorporates a Fabric Mezzanine Module (FMM), a technology at the core of Advantech's Customized COTS framework offering distinct differentiation advantages over COTS platforms and in-house designs.
Advantech also announced their FWA-6510 Network Appliance Series using two Intel® Xeon® processors E5-2600 with 16 physical cores and offering outstanding I/O density scaling up to sixty-four gigabit Ethernet ports or sixteen 10GbE ports in a modular architecture. Network Module offerings allow many port configurations with or without network bypass, offload and acceleration.
"A key aspect of workload consolidation on a single architecture is software reuse. With Advantech's scalable hardware platform ranging from appliances to high end ATCA systems customers have the foundation for greater efficiencies, lower time-to-market and reduced life cycle management costs" said Peter Marek. Director of x86 solutions at Advantech's Networks and Communications Group. "With a single-architecture running control and data plane workloads, the re-use of hardware and software components makes it easy to apply them to new network elements or different Intel processors. It also facilitates the implementation of adaptable and software configurable network elements."
"Faced with increasing CapEx and OpEx, service providers are looking to telecom and networking OEMs to deliver optimized cost/performance ratios and greater energy efficiencies" said Steve Price, Director of Marketing at Intel's Intelligent Systems Group. "Now they have a truly compelling architecture that offers scalable platform choices for consolidating workloads on Intel® architecture, which dramatically reduces development effort, power consumption and time-to-market."
MIC-5322 AdvancedTCA Blade
MIC-5332 ATCA blade enables the highest performance available in ATCA form factor with up to 16 cores and 32 threads of processing power, fast PCI Express gen.3 lanes running at up to 8Gbps, and best in class virtualization support. Two QPI interfaces between the CPUs improve memory and I/O access throughput and latencies when one processor needs to access resources hosted by the other socket. With four DDR3 DIMMs per socket in a quad channel design running up to 1600MT/s, the MIC-5332 not only offers superior memory bandwidth over 3-channel designs, but can also support memory densities up 256GB using latest LR DIMM technology. It outperforms previous generation dual socket designs while keeping similar thermal characteristics with balanced airflow resistance. Using the Intel® C600 series chipset with its integrated 4-port SAS controller, the need for an external storage controller is eliminated making the MIC-5332 an ideal choice for cost sensitive control plane applications.
The blade also introduces the company's first use of Fabric Mezzanine Module (FMM) technology providing expansion capabilities for front I/O or additional fabric ports at significantly lower cost and real-estate than AdvancedMCs. While the blade supports two 10GBaseKX4 interfaces in the base model, dual dual-star fabric implementations can be added by installing an FMM-5001 module. FMM type II sockets with PCIe x16 connectivity provides extension possibilities for additional front port I/O, offload and acceleration controllers such as Intel® QuickAssist(TM) accelerators, IPSec offload engines or customer specific logic. Moreover, FMMs can be reused on RTMs and across different blade designs. This unmatched flexibility combined with the highest performance Intel Xeon processors available make the MIC-5332 equally well suited for both application and data plane workloads.
FWA-6510 2U Rack mount Appliance
The FWA-6510 is a compact 2U rack mount form factor with redundant AC or DC power supplies, up to four 2.5" drives, management I/O and a LCD module. With multiple PCIe x8 pipes balanced between CPU sockets, the system supports up to 4 slots of I/O modules in Advantech's Network Mezzanine Card (NMC) form factor with an 8-slot version in development supporting as much as 160Gbps of IO traffic.
Deep Packet One Evaluation Platform
Based on the FWA-6510 160Gbps-capable 2U appliance, the Deep Packet One Evaluation Platform comes pre-loaded with the Intel® DPDK, the 6WINDGate(TM) software and the Qosmos ixEngine(TM) . The platform enables fast-track evaluation of the latest multi-core technology from Intel and the Intel® Embedded Alliance will allow customers to rapidly develop and test content driven network intelligence applications based on DPI technology. Deep Packet One was introduced to facilitate the technology evaluation process for OEMs by offering a turnkey evaluation platform based on the major building blocks required to evaluate DPI on Intel® architecture. Supported by Advantech's broad hardware offerings, OEMs can begin their development and evaluation work on this turnkey appliance and later move to other target platforms such as ATCA easily.
Charlie Ashton, VP Marketing, 6WIND
"Networking software from 6WIND delivers advanced packet processing functions that are critical for maximizing the performance of DPI-based products on Intel® architecture. With the Deep Packet One Evaluation Platform, developers will discover how multi-core packet processing with 6WINDGate and Intel® DPDK provide performance which scales linearly with the number of cores until the maximum bandwidth of the hardware platform is reached. Processor cores not used to run 6WINDGate are available to run value-added application software or Virtual Machines (VMs), resulting in a highly efficient and flexible system for advanced networking equipment."
Erik Larsson, VP Marketing, Qosmos
"Application developers can accelerate time-to-market by leveraging Qosmos ready-to-use DPI software on the latest Intel® Xeon® Processors. Advantech's Deep Packet One is an ideal platform for enabling the classification of traffic at core network speeds and extraction of metadata in real-time".
The MIC-5332, FWA-6510 and Deep Packet One Platforms are released products and available to order. For fully detailed product and ordering information, please visit the dedicated web page at www.advantech.com/...
