e-mesh

[joaofl]

I've been working with real-time networks for a while, and lately I'm specially focusing on mesh grid networks for NoCs.

To develop my research, I have built for the last years, a module for simulating mesh-grids at the protocol level, using ns3 (network simulator 3). The simulator is already working, and have given me good insights on the performance of such networks.

Right now, my objective is to implement the epiphany e-mesh architecture and protocol on my simulator, and compare results of performance measurement of the simulator and the chip itself.
But for that, I would need greater details of the e-mesh architecture and its protocols. The content provided in the documentation is too shallow, and it is not enough for me to implement it.

Can anybody give me an idea of where and how to have access to such information? This is part of my PhD research, and my objective with this is to build a proof of concept of a NoC simulator module for ns3.

The simulator is open source, and can be accessed here:

[aolofsson]

Moving topic to Q&A. Please post some examples of information that is missing.
Also, see the following discussion.
viewtopic.php?f=23&t=637&hilit=mesh

[joaofl]

Dear Olofsson, thank you for your reply.

I took some time to take a better look into the topic you suggested (about the arbitrations), and the patent of yours. I got some, and I'll probably get many more of the details I need there.

But what I still miss is more details about the protocol itself. For example, an overall perspective of what I'm trying to reproduce on my simulator:
Like in the "all-to-one" epiphany example in my parallella, In a write operation from all nodes to node 0,0, what happens in the network such that the total throughput measured is X?

In more details...:
The c-mesh does 64 bits atomic write operations, in 1.5 clock cycles per hop. If you want to write some several words, how would the packets look like? What is the protocol overhead?

Is the packet as simple as [ 12 bits for destination + 32 memory address (word aligned?) + 20 bits of data ] ?
Following packets look the same? If not, does the first one establish some virtual channel? Is it blocking?

The manual says that the r-mesh performs one read operation every 8 clock cycles. Is the read operation an 8 bytes atomic transaction as well?
Does It takes 8 clock cycles because the bandwidth is smaller, or because it has some sort of time division due some limited resource?

Does the packet look like: [ 12 bits for destination + 12 origin + 32 memory address (word aligned?) + 8 offset ] ?

Does it uses wormholing? What are the flit sizes? How does the flow control works, and the time it consumes? If not, how does it do queuing?

Sorry if these look like a silly question, but I believe the more details I understand, the closer the simulator can get to simulate the real architecture network performance. With that, better metrics can be taken and understood.

One good thing about ns3 is that it uses c++, and I could potentially use the same c files from an epiphany example, and run it on virtual nodes. I would be able to insulate processing time, and analyse the impact of communication (alone) in the overall performance.

Thanks again for your attention.

Best,

João

[aolofsson]

[joaofl]

Adreas, thanks again.

I also found useful information in the Epiphany Datasheet, and at the Git documentation (For example the table below @ https://github.com/parallella/parallell ... /src/elink).
Although I still could not find what are the various packet modes.

In this context, I would like to ask your personal opinion about the importance of such simulation tool for the NoC community. What would you expect to measure/test, and which kind of statistics would you look for?

The simulation runs already. Packets travel between nodes, although I still miss implementing some minor details, which are part of my ongoing effort now.

Thanks again.


[/td][/tr]