I agree with Tim and furthermore the community Pharos labs are underutilized. The Intel Pharos lab alone hosts about 72 servers of current generation Xeon which would be very expensive to replicate in a 3rd party hosted environment.
As a performance project VSPERF has had experience with hosting "special" hardware ... traffic generators, platform/processor skus, BIOS features, hardware configurations (e.g. NICs) and network setup are all important. This kind of setup ... new hardware with particular configurations (and usually issues) often requires some "hands on" debugging and effective support means good communication between the lab admins and the project. Throw SmartNICs and FPGA accelerators into the mix and I am skeptical that a large 3rd party hosting company will deliver. In many cases this cutting-edge hardware will be severely constrained and hard to justify sending off to a 3rd party lab. I think we are also underestimating the potential value of the Pharos labs (as an output of OPNFV vs a burden of sunk cost) to the companies that host them ... as a way of proving integration in the open source ecosystem (Tim's point). If we can offer the LaaS tooling and support the community Pharos labs with all our Infra activities and expertise IMO it would be huge value for companies internal purposes as well as for their open source objectives. For sure some synergy between LNF projects is needed on lab infrastructure ... we currently also host an independent ONAP lab ... and now Akraino, StarlingX etc. are trying to set lab infrastructures. /Trevor C -----Original Message----- From: [email protected] <[email protected]> On Behalf Of Tim Irnich Sent: Friday, December 7, 2018 1:32 AM To: Trevor Bramwell <[email protected]>; [email protected] Subject: Re: [opnfv-tech-discuss] Infra Evolution: Labs We seem to be forgetting that a significant fraction of our hardware comes through donated Pharos labs. I think it is important to maintain and evolve the possibility for companies to hook up labs with hardware of their choosing (within the limits of the Pharos spec) to the OPNFV federated labs infrastructure and run the OPNFV stacks on their hardware. Particularly in the light of the increasing importance of hardware accelerators, Pharos labs have the potential to enable OPNFV to become a proving ground for integration of such hardware into the broader open source ecosystem. This is a unique capability of OPNFV that, as far as I am aware, exists nowhere else in the industry. Tim On 12/7/18 1:47 AM, Trevor Bramwell wrote: > Hi all, > > The TSC tasked the Infra-WG to create a proposal for evolving the > infrastructure of OPNFV, but given the low attendance to the Infra-WG > calls I thought it best to bring this discussion to the mailing list so > we can benefit from the wider community's experience and knowledge. > > To keep the discussion focused, I'd like to just look at Labs and how we > can evolve the infrastructure there, and I'll follow up with thoughts on > CI, code, and artifacts, after we've reached some consensus since there's a > bit more interdependence with those components along with some > complexity. I've collected some of the options here[1]. > > What we need to find (and anyone feel free to jump in if I've > misunderstood the ask) is a scalable solution for bringing up new PODs > for CI that doesn't require us hosting or purchasing more dedicated > hardware. > > Now of the top cloud providers: AWS, Google Compute, Azure, only AWS > provides a hardware-as-a-service offering[2]. The others recently > enabled nested virtualization on some instance types, which may work for > us verifying virtual deployments (with performance trade-offs > obviously), but don't help when we want to verify against networking > hardware. > > This is why I started the discussion with Packet.net. They appear to be > the only provider that not only manages hardware for you, but provides > an API for accessing and provisioning it. > > What I'd like to know is if there are other options out there for > using/paying for hardware like this? Perhaps one of the other cloud > providers (IBM, HP, etc) have a service we haven't heard of? > Or are there other ideas of how we can manage this need? > > Happy to hear any and all input! > > Regards, > Trevor Bramwell > > [1] https://etherpad.opnfv.org/p/infraevolution > [2] https://aws.amazon.com/ec2/instance-types/i3/ > -- Dr.-Ing. Tim Irnich, Senior Program Manager Developer Engagement E-Mail: [email protected] Mobile: +49 172 2791829 SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#22520): https://lists.opnfv.org/g/opnfv-tech-discuss/message/22520 Mute This Topic: https://lists.opnfv.org/mt/28632364/21656 Group Owner: [email protected] Unsubscribe: https://lists.opnfv.org/g/opnfv-tech-discuss/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
