[P4] FW: [SIGCOMM '16 Posters and Demos] Paper #66 "High speed packet forwarding compiled..."

[Tejfel Máté]

Sziasztok!

   Szerintem egyelőre jó így, valószínűleg update-elni kellene, mielőtt 
nyilvánosabbá tesszük. Addig az UnivDay-es videó jobb PR-nak...

                M.

2016-06-09 00:33 keltezéssel, Leskó Dániel írta:
> Sziasztok!
>
>  A Sigcomm demo video-nak mi legyen a sorsa? Jelenleg egy teljesen 
> külön aloldalon van, link nem mutat rá sehonnan.
>
> Dani
>
> 2016.06.08. 7:10 keltezéssel, Sándor Laki írta:
>>
>> Fyi
>>
>> Sony Xperia™ okostelefonomról küldve
>>
>>
>>
>> ---- Eredeti üzenet ----
>> Tárgy: [SIGCOMM '16 Posters and Demos] Paper #66 "High speed packet 
>> forwarding compiled..."
>> Küldve: 2016.06.08. 0:15
>> Feladó: SIGCOMM '16 Posters and Demos HotCRP 
>> <noreply at sigcomm16posters.hotcrp.com>
>> Címzett: Sándor Laki <lakis at elte.hu>
>> Másolatot kap: 
>> marco.canini at uclouvain.be,daniel at land.ufrj.br,chesteve at dca.fee.unicamp.br,panhui at cse.ust.hk
>>
>> Dear Sándor Laki,
>>
>> Congratulations. The poster/demo entitled High speed packet forwarding
>> compiled from protocol independent data plane specifications that you 
>> have
>> submitted to SIGCOMM '16 Posters and Demos has been accepted. We received
>> 102 <tel:102> submissions and the TPC has accepted 39 of them.
>>
>> The reviews and TPC comments are attached below. We will send you an
>> additional email with instructions to prepare the camera-ready version of
>> your manuscript, which is due June 21.
>>
>> You can also access your reviews through the website:
>> https://sigcomm16posters.hotcrp.com/paper/66?cap=066atXs4RzoZ4G4
>>
>> Thanks again for having submitted your work to SIGCOMM '16 Posters and
>> Demos.
>>
>> - Marco, Daniel, Christian, Ben
>>
>> ===========================================================================
>>                  SIGCOMM '16 Posters and Demos Review #66A
>> ---------------------------------------------------------------------------
>> Paper #66: High speed packet forwarding compiled from protocol 
>> independent
>>            data plane specifications
>> ---------------------------------------------------------------------------
>>
>>                       Overall merit: 4. Accept
>>
>>                       ===== Comments for author =====
>>
>> This abstract describes a P4 compiler that translates to C using the 
>> Intel DPDK. Data-plane programmability in general, and P4 in 
>> particular are hot topics. However, there is a lack of open-source, 
>> publicly available compilers and tools available. This could be a 
>> valuable contribution, and an interesting demo.
>>
>> I would suggest you make a strong editing pass over the abstract to 
>> fix English and grammar mistakes.
>>
>>        SRC - Problem and motivation: 4. Good
>>
>> ===========================================================================
>>                  SIGCOMM '16 Posters and Demos Review #66B
>> ---------------------------------------------------------------------------
>> Paper #66: High speed packet forwarding compiled from protocol 
>> independent
>>            data plane specifications
>> ---------------------------------------------------------------------------
>>
>>                       Overall merit: 3. Weak accept
>>
>>                       ===== Comments for author =====
>>
>> The paper/poster shows the approach to develop a high performance P4 
>> (programming protocol independent packet processors) compiler.  The 
>> compiler works by seperating HW independent from HW dependent sections.
>> The compiler/demo will show that a P4 program (L2 switch, L3 router, 
>> etc) can be compiled to Intel DPDK enabled NIC and results in high 
>> performance.  However, the compiler/demo is currently limited to only 
>> Intel DPDK, which does not give evidence to the main goal of multiple 
>> targets.  Further, there are other P4 compilers that were not 
>> discussed from Xilinx (PX), Netronome, etc.  Further, it was not 
>> clear why a different hardware abstraction library (HAL) was needed 
>> and why the high level intermediate representation (HLIR) was not 
>> sufficient. Also, what was the tradeoff of using other intermediate 
>> representations to accomplish the same goals such as openCL, LLVM, etc.
>>
>>        SRC - Problem and motivation: 4. Good
>>
>>
>>
>>
>> _______________________________________________
>> P4 mailing list
>> P4 at plc.inf.elte.hu
>> https://plc.inf.elte.hu/mailman/listinfo/p4
>
>
>
>
> _______________________________________________
> P4 mailing list
> P4 at plc.inf.elte.hu
> https://plc.inf.elte.hu/mailman/listinfo/p4

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://plc.inf.elte.hu/pipermail/p4/attachments/20160609/720232a0/attachment.html>