The Linux Foundation is looking to boost access to real-time analytics with the adoption of the former Open Border Gateway Monitoring Protocol (OpenBMP) platform into the open source community under the new SNAS.io (Streaming Network Analytics System) initiative.
The program is set to provide a framework for the tracking and analyzing of network routing topology data in real-time for operators and enterprises using Border Gateway Protocol (BGP). The system taps into data collected from Layer 2 and Layer 3 of the network and includes IP information, quality of service requests, and physical and device specifications.
Arpit Joshipura, GM for networking and orchestration at the Linux Foundation, said SNAS.io is able to glean this data from the network in real time to allow DevOps, NetOps, and network application developers to better automate and manage their infrastructure and access data coming from the network.
Joshipura, who joined the Linux Foundation late last year, linked the need for increased visibility into real-time analytics to the ongoing move toward 5G, the Internet of Things (IoT), and greater use of software-defined networking (SDN), and network functions virtualization (NFV) across the communications space.
“There are a lot of tools to do the required number-crunching, but this requires automation in the form of something like ONAP,” Joshipura said, referencing the Linux Foundation’s recently created initiative from the merger of Open-O and open source ECOMP. “It’s the ability for those applications to make sense of the data that comes in and then to figure out what to do with that information in real-time. That’s the framework of what we are trying to do here.”
Contributors to the project include Cisco, the Internet Initiative of Japan (IIJ), Liberty Global, Pmacct, RouteViews, and the University of California, San Diego.
Serpil Bayraktar, principal engineer at Cisco, further highlighted the importance of real-time data capture and analysis embedded in SNAS.io in helping operators, enterprises, and application developers take advantage of the growing amount of data traveling across networks.
“The current model is to use telemetry to push data out of the router at specific intervals for data analysis, but this can also miss some data as it’s only near-real time and not real time,” Bayraktar said. “But in terms of the control plane and the routing of data, you are introducing risk opportunities. For SNAS.io, it becomes part of the system.”
Bayraktar also cited the scalability of what SNAS.io is looking to bring to the table, noting it can provide access to all data, with software-defined applications then being able to parse out the relevant information.
“On the collection side, someone needs to make sense of it,” Bayraktar said. “This has slowed work in this area because no one has been able to parse through that data. Our collector parses it and then produces data in a structured manner.”
In terms of how SNAS.io will fit into the Linux Foundation’s robust group of open source initiatives, Joshipura used a diagram to show that the program slides next to the current Platform for Network Data Analytics (PNDA) initiative, which is more focused on big data, with both PNDA and ONAP tapping into efforts coming out of SNAS.io. Cisco, in a statement, also referenced a link for SNAS.io to the Linux Foundation’s Fast Data Project (FD.io), Cloud Foundry, OPNFV, and OpenDaylight.
The real-time aspect of SNAS.io could allow the platform to remain a somewhat autonomous working group in the near term as Joshipura highlighted that difference when compared with PDNA. The Linux Foundation’s recent move to consolidate Open-O and ECOMP into ONAP showed it was willing to merge efforts that had a more competitive than complementary nature.
“If they compete they will be merged,” Joshipura said of how the organization may look toward further harmonization of its various initiatives. “If they are complementary they will be kept.”