mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 12:21:27 +01:00
4b2ae2ac33
This adds a new `Collectives` API for doing distributed collectives operations. This is intended to replace the [current Elastic store abstraction](https://github.com/pytorch/pytorch/blob/main/torch/distributed/elastic/utils/store.py) with more performant and debugable primitives. Design doc: https://docs.google.com/document/d/147KcKJXEHvk1Q6tISLbJVvLejHg_1kIhBQeu-8RQxhY/edit The standard implementation is using `StoreCollectives` but other more performant backends will be added in a follow up PR. Test plan: ``` python test/distributed/test_collectives.py -v ``` This tests both functionality using multiple threads as well as timeout behavior. Pull Request resolved: https://github.com/pytorch/pytorch/pull/125978 Approved by: https://github.com/shuqiangzhang
60 lines
1.5 KiB
C++
60 lines
1.5 KiB
C++
#pragma once
|
|
|
|
#include <condition_variable>
|
|
#include <mutex>
|
|
#include <unordered_map>
|
|
|
|
#include <torch/csrc/distributed/c10d/Store.hpp>
|
|
|
|
namespace c10d {
|
|
|
|
class TORCH_API HashStore : public Store {
|
|
public:
|
|
~HashStore() override = default;
|
|
|
|
void set(const std::string& key, const std::vector<uint8_t>& data) override;
|
|
|
|
std::vector<uint8_t> compareSet(
|
|
const std::string& key,
|
|
const std::vector<uint8_t>& expectedValue,
|
|
const std::vector<uint8_t>& desiredValue) override;
|
|
|
|
std::vector<uint8_t> get(const std::string& key) override;
|
|
|
|
void wait(const std::vector<std::string>& keys) override {
|
|
wait(keys, timeout_);
|
|
}
|
|
|
|
void wait(
|
|
const std::vector<std::string>& keys,
|
|
const std::chrono::milliseconds& timeout) override;
|
|
|
|
int64_t add(const std::string& key, int64_t value) override;
|
|
|
|
int64_t getNumKeys() override;
|
|
|
|
bool check(const std::vector<std::string>& keys) override;
|
|
|
|
bool deleteKey(const std::string& key) override;
|
|
|
|
void append(const std::string& key, const std::vector<uint8_t>& value)
|
|
override;
|
|
|
|
std::vector<std::vector<uint8_t>> multiGet(
|
|
const std::vector<std::string>& keys) override;
|
|
|
|
void multiSet(
|
|
const std::vector<std::string>& keys,
|
|
const std::vector<std::vector<uint8_t>>& values) override;
|
|
|
|
// Returns true if this store support append, multiGet and multiSet
|
|
bool hasExtendedApi() const override;
|
|
|
|
protected:
|
|
std::unordered_map<std::string, std::vector<uint8_t>> map_;
|
|
std::mutex m_;
|
|
std::condition_variable cv_;
|
|
};
|
|
|
|
} // namespace c10d
|