Re: Ubergraph - request for volunteers to benchmark alternative implementation

Just tried it out and the improvements were substantial. I tried to add 94K nodes (journal entries) to a graph, most of which have multiple edges, and with 0.7.2 it took 137 seconds, whereas with the specter branch, it only took 92 seconds. Everything else worked fine as well. Thanks for this l

Re: Ubergraph - request for volunteers to benchmark alternative implementation

Just tried it out and the improvements were substantial. I tried to add 94K nodes (journal entries) to a graph, most of which have multiple edges, and with 0.7.2 it took 137 seconds, whereas with the specter branch, it only took 92 seconds. Everything else worked fine as well. Thanks for this l

Re: Ubergraph - request for volunteers to benchmark alternative implementation

That's great news. Thanks for the report. On Sat, Aug 17, 2019 at 7:59 AM Matthias Nehlsen wrote: > Just tried it out and the improvements were substantial. I tried to add > 94K nodes (journal entries) to a graph, most of which have multiple edges, > and with 0.7.2 it took 137 seconds, whereas w

Ubergraph - request for volunteers to benchmark alternative implementation

I didn't strip it down for a true benchmarking test and my use case is large but fairly sparsely connected graphs, but I get about 10% improvement. -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegro

Re: Ubergraph - request for volunteers to benchmark alternative implementation

Thanks. On Sat, Aug 17, 2019 at 2:41 PM Colin Taylor wrote: > I didn't strip it down for a true benchmarking test and my use case is > large but fairly sparsely connected graphs, but I get about 10% improvement. > > -- > You received this message because you are subscribed to the Google > Groups