| | 45 | ==== Community Agreement from the Meeting ==== |
| | 46 | The session concluded with community agreement on three points: |
| | 47 | * The community will work to agree on common semantics |
| | 48 | * Ilia has proposals for layer 2 network resources and compute resources, and has recruited community members for other areas; thanks in advance to Max Ott, Hongwei Zhang, Martin Swany, Mike Zink, and Rob Sherwood. These individuals, and others who volunteer to help, will produce ontologies to represent the agreements. |
| | 49 | * GENI will use ProtoGENI RSpec V2 as the format on the wire |
| | 50 | * New concepts will be represented as PG RSpec extensions, allowing free addition and evolution of new concepts as needed. |
| | 51 | * Aggregates can use Translators to convert between formats |
| | 52 | * Ilia is building one between PG and Orca formats. |
| | 53 | |
| | 54 | ==== Next Steps Agreed to at the Meeting ==== |
| | 55 | For all of these, community involvement is key. Email dev@geni.net or Aaron Helsinger (ahelsing at geni.net) to get involved. Results will be published to dev@geni.net and to the GENI wiki to keep everyone informed of progress. |
| | 56 | ===== Agree on Concepts ===== |
| | 57 | * Align edge compute resource concepts, starting with Ilia's proposal. |
| | 58 | * 4 key semantic extensions |
| | 59 | * Wireless (Max Ott, Hongwei Zhang) |
| | 60 | * Measurement (Martin Swany) |
| | 61 | * Storage - physical & cloud (Mike Zink) |
| | 62 | * !OpenFlow flowspace (Rob Sherwood) |
| | 63 | * Produce ontologies for possible ratification at GEC11 |
| | 64 | * The same groups will produce and present these agreements |
| | 65 | |
| | 66 | ===== Adopt Schemas ===== |
| | 67 | * Once the semantics of these things are agreed to, Jon Duerig, Ilia and the GPO will assist in representing these as PG extensions |
| | 68 | * Revisit PG schemas |
| | 69 | * Change more attributes to be elements as needed |
| | 70 | * Make some types less opaque as needed |
| | 71 | * Ensure base is extensible for the known needed extensions |
| | 72 | * Jon Duerig of ProtoGENI will handle this |
| | 73 | * Once adopted, RSpec schemas will be maintained by the community, along with the AM API. |
| | 74 | * Modify the AM API specification to require that RSpecs are in the PG RSpec V2 format |
| | 75 | * Aaron Helsinger and the GPO will handle this, when the support infrastructure is in place |
| | 76 | ===== Implement Translators ===== |
| | 77 | * Write and deploy translators as necessary (coordinate with Ilia for a start - ibaldin at renci.org) |
| | 78 | * Enable conversion as a GENI-wide service |
| | 79 | * Replicating Ilia's converter as necessary |
| | 80 | * Test full PG/Orca interoperability by GEC11 |
| | 81 | * Build a thin PL/PG translation layer at PL (Andy Bavier, Tony Mack, Jon Duerig) |
| | 82 | * Other translators as necessary |
| | 83 | |
| | 84 | ==== Selected Discussion Points from the Meeting ==== |
| | 85 | ===== Ilia Baldine ===== |
| | 86 | * Translators may not be a long term solution, but do provide a way to make progress. |
| | 87 | * The key requirement is that we agree on semantics. |
| | 88 | * For networking, NDL is a superset of PG, which is a superset of PL. GENI primarily works at the PG/PL level, and translators can work there. |
| | 89 | * Networking semantics will deal with ethernet only for now - leave lower levels like optical for Orca or others with interest. |
| | 90 | * Adaptations are a key concept - e.g. part of describing stitching capability. |
| | 91 | * RENCI will maintain and extend their converter for PG/Orca formats. |
| | 92 | * Ilia described a proposal for a compute resources ontology. |
| | 93 | * We should draw on work in the cloud community. |
| | 94 | * The Open Virtualization Format (OVF) standard from the Distributed Management Task Force (DMTF) may be a good source for VM representations. |
| | 95 | * IETF Common Information Model (CIM) standard is also relevant, but is very complicated. |
| | 96 | |
| | 97 | ===== Martin Swany ===== |
| | 98 | * Uniform Network Information Schema (UNIS) from U Delaware leverages NML, perfSonar. |
| | 99 | * Philosophical points |
| | 100 | * Use the same schema for control & measurement |
| | 101 | * Express common stuff in a common way |
| | 102 | * Not everything is common |
| | 103 | * Non common stuff goes in an extension |
| | 104 | * Express complexity only as necessary |
| | 105 | * RFC6020 - YANG - provides an interesting new data modeling language. |
| | 106 | * Private or novel concepts should be kept separate with different namespaces. |
| | 107 | * N^2^ translators should be a concern: a good common internal representation inside the translator reduces this problem. |
| | 108 | |
| | 109 | ===== Rob Ricci ===== |
| | 110 | * Maximize tool support: any programming language or environment should be supported. |
| | 111 | * Use an extensible format using namespaces |
| | 112 | * Add annotations as needed |
| | 113 | * The maintainer of the format must not be a bottleneck to innovation |
| | 114 | * The RSpec schema maintainer could be GENI |
| | 115 | * Projects can describe new stuff with a new extension without waiting for any kind of global consensus. |
| | 116 | * Format must have a defined base that evolves over time. |
| | 117 | * Bound the # of active versions of the schema at 2. |
| | 118 | * Services advertise which formats they support, and which extensions. |
| | 119 | * 'Consenting Adults' model for extensions |
| | 120 | * Those who want the extension agree at first & only they understand it. |
| | 121 | * Eventually the extension may get 'blessed', to avoid duplication of concepts. |
| | 122 | * PL vs PG format: PL has a 'site' tag, which is similar to OF 'network'. |
| | 123 | * Otherwise, it is quite similar to the PG format. |
| | 124 | |
| | 125 | ===== Rob Sherwood ===== |
| | 126 | * OF extension would likely be very simple |
| | 127 | * May extend link types |
| | 128 | * Need to decide whether flowspace is defined relative to links, switches, or other. |
| | 129 | |
| | 130 | ===== Group ===== |
| | 131 | * Ilia: first step must be a UML diagram or similar, not the schema. No one can read XSDs |
| | 132 | * Andy Bavier |
| | 133 | * the proposal is fine. |
| | 134 | * PL will translate. |
| | 135 | * Martin Swany |
| | 136 | * this is like old UNIS, so I support this |
| | 137 | * Use XML elements not attributes to make extension easier. |
| | 138 | * RSpec ontology & schema should be co-hosted with the GENI AM API |
| | 139 | * Community maintains the schema |
| | 140 | * Perhaps change the namespaces and URLs to a geni.net address as extensions and base are 'blessed'. |
| | 141 | |
