Fujitsu Laboratories Ltd. today announced the development of technology to accelerate virtual routers, which play a
major role in the functionality of networks in virtual environments.Figure 1: High Speed Packet Addressing Control TechnologyFigure 2: Hybrid Memory Management Technology
Server virtualization, which concentrates multiple applications and the infrastructure functionality of network
processing in a general-purpose server environment, is spreading beyond datacenters to the field of edge computing,
including wireless base stations and Mobile Edge Computing (MEC)(1). The technology also proves useful in areas
including societal infrastructure with applications like traffic management, and in entertainment, where it can be used
to deliver viewers content like sporting events. As data volumes increase and systems become increasingly complex,
however, the CPU resources required for packet processing in a virtual network increase. This reduces the number of
applications that can run on a single server, and leads to lower server aggregation rates.
To resolve this issue, Fujitsu Laboratories has now developed technology to accelerate packet address control, which had
been a performance bottleneck, while also offloading the processing of router functions in the virtual network to field
programmable gate arrays (FPGAs)(2). This speeds up packet processing performance eighteen-fold compared with existing
virtual routers, while reducing the use of CPU resources to about one thirteenth that of existing technology.
With this technology, Fujitsu Laboratories will deliver high performance virtual networks with a low computational
burden, supporting the digital transformation of companies by improving server utilization efficiency for businesses
that will use large volumes of data in the 5G era.Development Background and Issues
Server virtualization, which concentrates multiple applications and network processing functionality using a virtual
environment on a general-purpose server, is becoming increasingly common and widespread as digital transformation
progresses. Moreover, by accelerating digital transformation within a wide variety of companies, new services are also
being created that link applications together. Against this background, the volumes of data handled by these
applications is trending upwards, leading to issues of increasing network complexity and increased computational
burdens.
As networks grow increasingly complex and computational burdens increase, issues have arisen with improving packet
processing performance in the virtual routers that are essential to flexible network structures, as well as with
reducing the CPU resources used for network processing, in the effort to deliver efficient server virtualization. By
resolving these issues, it will become possible to increase the number of applications that can run on a server, as well
as decrease the number of servers needed for a system, reducing investment costs for customers.About the Newly Developed Technology
Fujitsu Laboratories has now developed technology to offload the burden of virtual router functionality from the CPU to
FPGAs, alongside technology to increase the speed of packet addressing control. This will not only increase network
performance, but will also significantly reduce the CPU resources used for network processing.1. High Speed Packet Addressing Control Technology
Previously, high-speed processing was conducted by subjecting all packets to the same processing, without changing the
order of the packets. If the order of the packets were to change in the input or the output of the virtual router
processing, that might lead to applications resending packets, which would lower quality and increase the burden on the
system.
With this technology, packet sequence information is stored separately from the packets in the pipeline. Consequently,
in address search processing spread over multiple stages, when the results of an earlier search render subsequent
searches unnecessary and a packet bypasses those searches, it can be rapidly recombined with the other packets in the
pipeline in the proper order based on the packet pipeline sequence information. Combined with the pipeline duplication,
this technique reduces memory access while maintaining high pipeline processing speeds. These technologies improve
packet processing performance.2. Hybrid Memory Management Technology of Lookup Tables
Fujitsu Laboratories has developed technology to automatically switch between the high-speed, low-capacity memory within
the FPGAs and the high-capacity external memory, without pausing address search processing. In address search
processing, multiple tables are searched to determine the address. As the number of connections increases, the amount of
memory used by the lookup tables increases and the free space in the internal memory becomes insufficient. When this
happens, this technology can synchronize lookup tables with relatively low access frequency, relative to their size,
with external memory in the background, automatically switching to lookup tables in external memory without stopping
search processing. This enables stable packet processing performance, limiting access to external memory, even in cases
such as large-scale systems, where there are many communication destinations, and large lookup tables become essential.Effects
An evaluation of the effects of this offloading on a general-purpose server was conducted using this technology, which
was deployed on Intel Stratix 10 MX FPGA featuring high-speed HBM2(3) memory using Tungsten FabricTM, an exemplary
open-source virtual router. Two servers connected with 100Gbps ethernet ran four virtual machines each, and
communications were conducted between each virtual machine to test the performance of the virtual routers. The results
showed that packet processing performance, which was 13.8Mpps using existing methods, was increased to 250Mpps, or about
an 18-fold increase. In addition, the number of CPU cores used was reduced from 13 cores to 1 core.
Using this technology, it is now possible to increase application server density. This will support the digital
transformation of customers from the foundational level, improving server utilization efficiency in infrastructure
business areas that use large volumes of data in the 5G era, particularly in base stations and MEC for carrier
businesses, such as, for example, making it possible to operate a video distribution service for a stadium with fewer
servers.Future Plans
Fujitsu Laboratories will continue to improve the performance and expand the functionality of this technology, as well
as conducting further evaluations based on customer digital transformation use cases, with the goal of commercialization
during fiscal 2021.
(1) Mobile Edge Computing (MEC)
Technology primarily used in mobile communications networks that delivers reduced latency and a more distributed burden
for core networks by conducting data processing using servers and storage distributed to locations such as base
stations.
(2) Field programmable gate arrays (FPGAs)
Field programmable gate arrays are large-scale semiconductor devices for which logic circuits can be rewritten.
(3) HBM2
The memory that is built into the same package as the FPGAs which is capable of transfer data at high-speed and
high-capacity.