Disaggregation has been considered as the main concept of next-generation open network architecture. Recent trends of softwaredefined network (SDN) and network function virtualization (NFV) have performed successfully in bringing up open and interoperable elements into existing closed-architecture network. However, to reach the ultimate goal of “open” network, the importance of hardware should not be overlooked and in fact, the processor design and the hardware/middleware API libraries play a significant role in the transition to total open network. This paper will discuss the challenges faced in today’s network structure and how the hardware will play a critical role to complement the open software.
Our white papers explore key areas of Lanner's expertise in developing and improving cutting-edge features for network computing applications.
According to recent researches, streaming videos have taken up a larger and larger share of the network traffic. It is predicted that videos may occupy almost 80 percent of the traffic at the time of this writing. That means, users of desktops, laptops and mobile devices are demanding higher quality of streaming videos. To address the demand, Intel® has launched the 6th Generation Core™ processors (codenamed “Skylake”), which integrates ultra processing power in graphics and data. The promise of the new processors will save server makers the TCO (Total Cost of Ownership) by delivering improved graphic streaming quality without the space-taking, powerconsuming PCI-e based graphic cards. On the other hand, servers powered by the 6th Generation Intel® Core™ improve serviceability at the competitive edge.
The increasing number of malicious web attacks has driven major corporations to have deployed multiple security measures including proxy servers, firewalls, multiple anti-virus scanning engines and encryption technologies against various types of cyber crime. While these measures are working 24/7 to keep the enterprise network safe, they system loads may eventually experience performance degradations due to high CPU workloads.
The drive to redefine enterprise total cost of ownership (TCO) has contributed to the rise of midrange systems, the type of servers that can process large-scale instructions and applications, though less powerful than mainstream servers. Midrange network servers have been favored by small-and-medium enterprises and regional division and subsidiaries of global corporations due to their affordable TCO, lower power consumption and scalable open architecture mainly by Intel architecture-based platforms.
Many of today’s networking devices are built with LAN bypass function to prevent failed LAN transmission from impeding important in-flow critical data. The function is to ensure fault-tolerance for network appliances so that essential communications an be still continued even when software, hardware or power failure occurs. The basic idea of bypass is “cross-connection”. When an Ethernet port fails, the dysfunctional port will be temporarily disconnected from the system interface and all the in-line data will flow into the other LAN port so that data is still received.
Secure Boot is a form of verified booting technology which ensures boot path validations. As defined in UEFI (Unified Extensible Firmware Interface) specifications, this BIOS-based technology assures that the system firmware checks if the boot loader is signed with a cryptographic key contained in the database in the firmware. In other words, Secure Boot is a firmware-based boot path validation mechanism that contains cryptographic key and it will check if the boot loader is protected by the key. Only the ones with proper digital signature verification in the next-stage boot loaders, kernels or user space can access the system. This will prevent the execution of codes that are not signed with the cryptographic key programmed in the system firmware.
Intel® DPDK is a software set of libraries and Ethernet drivers compatible on any Intel x86 processors for accelerated packet processing. It is designed to run in Linux environment for enterprise and industrial network management.
The libraries of Intel DPDK improve data plane performance, pre-fetch data, trim memory latency down and reduce development time so that enterprises can save tremendous cost in vertically-integrated and monolithic hardware such as evolved packet cores. On the other hands, with software assistance, the platform can be easily scalable and more flexible since packet processing and other workloads are executed on Intel processors and DPDK. This indeed will help businesses move to SDN (Software Defined Network) and NFV (Network Function Virtualizations) without excessive ownership cost and development time.
The launch of Intel® Xeon® processor E5 v3 series platform indicates that IT management is heading into a new generation of processing capabilities. The new Intel® Xeon® server-based platform presents a great leap in CPU performance, RAM efficiency, PCH connectivity, and networking modules to deliver unprecedentedly low latency, high bandwidth and relatively lower power consumption.
By integrating Intel® Xeon® v3 processors, DDR4 memory, the most up-to-date Intel® QuickPath Interconnect (Intel® QPI) , Intel® Virtualization (Intel® VT) technologies, and the unparalleled networking Intel® Ethernet XL710 family (codenamed Fortville), the Intel® Xeon® processor E5 v3-based server platform (codenamed Grantley) is architected to address today’s data center challenges.
The powerful coupling of the new Intel Xeon processor E5-2600 v3 and the Intel C610 PCH, codenamed “Grantley,” is a platform designed to meet the challenges of today’s virtualization and security management.
Improvements offered by the Grantley platform are evident in Lanner's FW-8896, a 2U rackmount network appliance capable of boosting network appliance’ processing power, memory bandwidth, Ethernet throughput and I/O efficiencies for consolidating virtual machines, conducting SDN and for managing heavy network traffic.
In this whitepaper, we will demonstrate the performance differences between Intel new Intel Xeon processor E5-2600 v3 and previous generations by conducting benchmark tests on Lanner products.
In the cloud computing era people are connected to the Internet through their devices wherever they go. However, with all this convenience of data accessibility and availability come big challenges in data protection.
Data protection is one of the most important security issues on the Internet. Using security features to protect data, applications, and systems in the cloud infrastructure is critical in both preventing data leakage and in managing compliance mandates for service providers.
To help reach this goal efficiently and cost-effectively, Lanner Electronics has developed security acceleration adaptors and enterprise-class servers based on different SKUs of the Intel® Communications Chipset 89xx Series. These appliances ensure high security and reliability in a virtualized environment for quick deployment and rapid delivery of cloud services.
In this whitepaper, we will show the precise performance differences between various Intel Communications Chipset 89xx Series SKUs by conducting benchmark tests on Lanner products based on this technology.