As open source software moves into the mainstream of telco platforms (see separate item), so must white box, commoditized hardware, which drives the economics of software-defined networks (SDN). That white box approach will drive new suppliers of chips and boxes, as well as a rising interest in open source hardware, even for the base station.
AT&T has been one of the biggest drivers of telco SDN and virtualization on the software side, and has shaken up some of its supply chain to support its massive Domain 2.0 project. However, for the physical networks, it has proved hard to reduce reliance on the traditional vendors, even if they have to fight to retain such dominant shares of AT&T’s deployments as in the end-to-end past. But now the US operator is putting those suppliers on notice too, even in hardware, with a move into white box switches that could sound a deathknell for closed, expensive equipment.
This is not just about cost for AT&T. Describing its switch at the Open Networking Summit last week, its CTO, Andre Fuetsch, highlighted the use of the open source pP4 programming language on the merchant silicon powering the white box device. Using P4, and tying the white box to the Open Network Automation Platform (ONAP), enables visibility down to a packet level, which could drive new types of services and quality mechanisms.
“This is more than just about lowering cost and achieving higher performance,” he said in his keynote. “Frankly that’s table stakes. This is really about removing barriers, removing layers, removing all that internal proprietary API stack that we’ve lived with these legacy IT systems, now we can bypass all of that and go straight to ONAP” to achieve fine-grained per-packet visibility.
AT&T has worked with Barefoot Networks and SnapRoute on its switch, which is now in field trials, and the presence of these small vendors will send chills down the spines of its traditional suppliers. Barefoot provided its Tofino chip and surrounding software, while SnapRoute (in which AT&T is an investor) developed the network operating system, Flex Switch.
Fuetsch said the firm “took this very novel approach of building a hardware abstraction layer and running open source networking modules like BGP and OSPF on top”. This layer can then operate independently of the silicon, and AT&T said it has other white boxes, based on different chips, in the works, with the same network OS. The second design to be moving into field trials is a white box from Delta Electronics, running on a Broadcom switch-chip.
“In December we got a Barefoot chip from the foundry and in less than three months had live network traffic running on it. I’ve never seen us deliver from chip to network production so quickly. That’s the power of white boxes,” he said.
In an interview, Fuetsch laid down the gauntlet to the major vendors, saying: “Here’s the big message for the OEMs. It is really a call for them to open up their architectures, open up their software and their own hardware so they can participate. They are going to have to make a choice here – do you want to be at the table or on the plate?”
Some OEMs are starting to make the transition. Cisco and Juniper are both using merchant chips, mainly from Broadcom, for some switches after years of designing hteir own ASICs, while Cisco was recently reported to be on the cusp of offering its software standalone, to run on any white box hardware. That move has already been made by some smaller players, such as Arista, which now offers its EOS software standalone.
However, some believe market dominance will move down the stack from the box makers to the chip suppliers. Broadcom dominates the white box space, whether for hyperscale cloud platforms or the nascent telco space, and its chips are used in 90% of Ethernet switches of any kind.
However, there are various start-ups hoping to ride the coming disruption – as well as Barefoot, Innovium recently announced its Teralynx family. Broadcom’s rivals will be keeping a sharp eye on such firms – indeed, Cavium acquired XPliant, which has reported design wins with Arista and Brocade. Also in the fray are Nephos, a spin-out from Mediatek, Marvell, Mellanox and China’s Centec Networks.
The white box model will only reach its maximum potential when it is supported by open source hardware and open APIs. Many are calling for open APIs to help operators to switch Ethernet switches in and out of their networks in the same way as they do with servers. And those APIs would lower the barriers to alternative chip providers to compete against Broadcom.
Unsurprisingly, Google is one of the loudest voices – partly because, as a massive buyer of white boxes and silicon, it would benefit from a competitive ecosystem; partly because it has a bigger interest in driving and defining cloud and network standards to suit its business models and heighten its influence.
“Right now there’s no standard API for switches so when plug in a new switch, we do that work ourselves,” Amin Vahdat, technical lead for networking at Google, told EETimes at the Open Networking Summit. “That really holds back the market.”
The company is putting its money where its mouth is, aiming to bring together a group of chip suppliers and customers to define a new interface within a year from now.
Open source, software-defined hardware will not remain confined to the switches and servers, but is starting to reach its tentacles right out into the radio network and the base stations.
While Intel and others have been pushing standard processors into base stations to support distributed cloud services and edge-based intelligence, the real disruption would come from an open software-defined radio (SDR). This is what Lime Microsystems is targeting with its high profile LimeSDR and LimeNET crowdfunded initiatives.
Lime claims it is the first company to take SDR into base stations, with the potential to slash the cost of building a RAN. This will become critical because of two trends which will require a new RAN cost base – densification, using large numbers of small cells; and the wireless IoT, with its need for ubiquitous coverage, but its low ARPUs.
Run in conjunction with Ubuntu creator Canonical, LimeNET is the second crowdfunding campaign, and aims to take the technology funded under the earlier LimeSDR initiative into carrier class macro and small cell base stations. Meanwhile, LimeSDR is the first $1m project under the CrowdSupply platform.
The UK company says it has “the potential to completely transform the way telco networks run, shifting the emphasis and value away from proprietary hardware to open hardware with app stores on top.”
Lime CEO Ebrahim Bushehri said in a statement: “Radio access technology for wide area networks accounts for a significant portion of the overall deployment cost. We’re seeking to make the hardware an open source commodity sold for a fraction of current offerings, with the real value being in the software it runs. Doing it this way would effectively turn LTE, GSM or LoRa, or even 5G, into just an app.”
The LimeNET Mini is an app-enabled, small cell base station that combines the LimeSDR technology with Intel’s x86 motherboard. It is targeted at localized IoT equipment, while the more-powerful LimeNET is for carrier-class wide area networks. The first operator to use the LimeSDR platform is EE, which is partnering with several UK universities to develop low cost base stations for remote areas.
Other applications developed by the LimeSDR community include IoT gateways, aviation transponders, smart meters, and systems for media streaming, radio astronomy, radar, drones and others.
For LimeSDR, the RF company provided its LMS7002M RF transceiver, which allows the board to support a continuous frequency range from 100 Hz to 3.8 GHz, meaning it can be programmed to work with most mainstream wireless technologies including UMTS, GSM, LTE and WiFi, plus sub-1 GHz connections like Sigfox and LoRa, and personal area networks like Bluetooth, ZigBee and Z-Wave. The chip supports 2×2 MIMO.
Much of the design of the board is open source, with the board schematics and layout available under a CC BY 3.0 licence, as well as the USB link and host software, and Altera’s Project Quartus software.
Under the Canonical partnership, developers will be able to write and publish applications inside the Snappy app store, where users can download them and then run them on their LimeSDR kits. Of course, developers are able to charge for these applications, with Canonical receiving a cut of that sale, but it seems likely that an ecosystem of free and open source apps will spring up around the LimeSDR.
Bushehri said at the launch last summer: “We are really excited about this App Store announcement with Snappy Ubuntu. Already we are seeing applications being posted in forums and this development goes a long way to ensure that wireless network and connectivity will be app-enabled”.
The LimeSDR will be supported by Lime’s Myriad-RF community, which it set up as a family of open source hardware and software projects for wireless innovation.