[Breaking] Switch from rabit to the collective communicator (#8257)

* Switch from rabit to the collective communicator

* fix size_t specialization

* really fix size_t

* try again

* add include

* more include

* fix lint errors

* remove rabit includes

* fix pylint error

* return dict from communicator context

* fix communicator shutdown

* fix dask test

* reset communicator mocklist

* fix distributed tests

* do not save device communicator

* fix jvm gpu tests

* add python test for federated communicator

* Update gputreeshap submodule

Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
This commit is contained in:
Rong Ou
2022-10-05 15:39:01 -07:00
committed by GitHub
parent e47b3a3da3
commit 668b8a0ea4
79 changed files with 805 additions and 2212 deletions

View File

@@ -1,16 +1,16 @@
#ifndef XGBOOST_TESTS_CPP_COMMON_TEST_QUANTILE_H_
#define XGBOOST_TESTS_CPP_COMMON_TEST_QUANTILE_H_
#include <rabit/rabit.h>
#include <algorithm>
#include <string>
#include <vector>
#include "../helpers.h"
#include "../../src/collective/communicator-inl.h"
namespace xgboost {
namespace common {
inline void InitRabitContext(std::string msg, int32_t n_workers) {
inline void InitCommunicatorContext(std::string msg, int32_t n_workers) {
auto port = std::getenv("DMLC_TRACKER_PORT");
std::string port_str;
if (port) {
@@ -28,12 +28,11 @@ inline void InitRabitContext(std::string msg, int32_t n_workers) {
return;
}
std::vector<std::string> envs{
"DMLC_TRACKER_PORT=" + port_str,
"DMLC_TRACKER_URI=" + uri_str,
"DMLC_NUM_WORKER=" + std::to_string(n_workers)};
char* c_envs[] {&(envs[0][0]), &(envs[1][0]), &(envs[2][0])};
rabit::Init(3, c_envs);
Json config{JsonObject()};
config["DMLC_TRACKER_PORT"] = port_str;
config["DMLC_TRACKER_URI"] = uri_str;
config["DMLC_NUM_WORKER"] = n_workers;
collective::Init(config);
}
template <typename Fn> void RunWithSeedsAndBins(size_t rows, Fn fn) {