Intel has unveiled an Intel Network Platform that provides the system-level reference architectures, drivers, and software building blocks needed to accelerate the rate at which 5G networking services will be rolled out.
Unveiled during an online preview of the discussions Intel hopes to lead during the Mobile World Congress 2021 event next week, the Intel Network Platform is part of a larger effort to accelerate deployment of applications at the point where data is increasingly being created, consumed, and analyzed, said Dan Rodriguez, corporate vice president for the Network Platforms Group at Intel.
As part of that effort, Intel is making available Intel Smart Edge, a toolkit developers employ to manage access to edge computing services, in two flavors. One is an open source Intel Smart Edge Open toolkit formerly known as OpenNESS, while the other is a commercial instance of the toolkit supported by Intel.
Finally, Intel is also expanding the Intel Agilex field-programmable gate arrays (FPGAs) to include support for integrated cryptography to better support security in addition to making available an Ethernet adapter designed for space-constrained locations.
Building Integrated Infrastructures
Intel sees the rise of edge computing as an opportunity to convince organizations that rely on proprietary ASICs or rival communications platforms from Broadcom to standardize on a mix of processors and software from Intel — making it easier to converge compute, storage and networking. As organizations process and analyze data at the point where it is created and consumed, the need for more integration at the infrastructure level becomes more apparent, noted Rodriguez.
Intel is betting that, in a few years, large workload volumes will be processed outside of a traditional data center or cloud computing environment.
“Analysts tell us 75% of workloads will be running outside of a data center,” said Rodriguez.
That doesn’t mean there will be fewer workloads running in the cloud or an on-premises data center. What it does mean is that the relationship between workloads spanning everything from the edge to the cloud and everything in between will become more nuanced. More challenging still, there’s no such thing as a single edge computing platform. Depending on the use case, an edge computing platform can be anything from a simple gateway to a complete hyperconverged infrastructure (HCI) platform. Each platform may be connected over a dedicated network connection or increasingly wireless 5G or even, one day, possibly a 6G connection.
Refining Data Management
In the meantime, IT teams may want to gear up for a new era of IT. Rather than data always being brought to compute in the cloud or a local data center, it’s clear computing is being brought to where data is being created and consumed. That shift will naturally have a major impact on how data is managed and secured as the amount of raw data moving across a wide area network (WAN) continues to decline. Of course, the amount of data being generated continues to grow unabated. The challenge now is figuring out where best to process that data based on latency requirements of any given application.