improve performance

[deflaux]

This package is in active development and there are some inefficiencies in a few places (e.g., the code that accumulates multiple pages of variants and reads). But we'll tackle improving performance a bit when the code settles down a bit more.

[pgrosu]

No worries Nicole, it's a team effort - that's what makes it fun :)

[siddharthab]

Going forward, performant access to the API is predicated by using the RPC interface instead of the REST interface:
http://www.grpc.io/
https://cloud.google.com/genomics/reference/rpc/

I am working on writing a running example based on gRPC.

[ttriche]

Does this mean that Avro is getting canned? And an R gRPC client coming?

AWESO^H^H^H^H^Hthat is an eminently logical implementation decision.

Statistics is the grammar of science.
Karl Pearson http://en.wikipedia.org/wiki/The_Grammar_of_Science

On Mon, Jul 6, 2015 at 4:31 PM, Siddhartha Bagaria <notifications@github.com

wrote:

Going forward, performant access to the API is predicated by using the RPC
interface instead of the REST interface:
http://www.grpc.io/
https://cloud.google.com/genomics/reference/rpc/

I am working on writing a running example based on gRPC.

—
Reply to this email directly or view it on GitHub
#17 (comment)
.

[siddharthab]

I am trying to base the gRPC R package on the RProtoBuf package. But there are several changes needed there too. A generic package might be a little farther off but I can get the GoogleGenomics endpoints running soon-ish.

[pgrosu]

Sid, how about just basically writing a wrapper via Rcpp? Autonomy is always more fun :)

Tim, I think it's an evolution :)

[siddharthab]

Paul, that is what will happen if I base it on the RProtoBuf package. I need the package for the interface to the C protocol buffers library.

[pgrosu]

Gotcha! Super-excited to see what you come up with :)

Thx,
~p

[siddharthab]

Resolved with #56