kvcache-ai-ktransformers/archive/ktransformers/server/utils/multi_timer.py
Jiaqi Liao 57d14d22bc
Refactor: restructure repository to focus on kt-kernel and KT-SFT modulesq recon (#1581)
* refactor: move legacy code to archive/ directory

  - Moved ktransformers, csrc, third_party, merge_tensors to archive/
  - Moved build scripts and configurations to archive/
  - Kept kt-kernel, KT-SFT, doc, and README files in root
  - Preserved complete git history for all moved files

* refactor: restructure repository to focus on kt-kernel and KT-SFT modules

* fix README

* fix README

* fix README

* fix README

* docs: add performance benchmarks to kt-kernel section

Add comprehensive performance data for kt-kernel to match KT-SFT's presentation:
- AMX kernel optimization: 21.3 TFLOPS (3.9× faster than PyTorch)
- Prefill phase: up to 20× speedup vs baseline
- Decode phase: up to 4× speedup
- NUMA optimization: up to 63% throughput improvement
- Multi-GPU (8×L20): 227.85 tokens/s total throughput with DeepSeek-R1 FP8

Source: https://lmsys.org/blog/2025-10-22-KTransformers/

This provides users with concrete performance metrics for both core modules,
making it easier to understand the capabilities of each component.

* refactor: improve kt-kernel performance data with specific hardware and models

Replace generic performance descriptions with concrete benchmarks:
- Specify exact hardware: 8×L20 GPU + Xeon Gold 6454S, Single/Dual-socket Xeon + AMX
- Include specific models: DeepSeek-R1-0528 (FP8), DeepSeek-V3 (671B)
- Show detailed metrics: total throughput, output throughput, concurrency details
- Match KT-SFT presentation style for consistency

This provides users with actionable performance data they can use to evaluate
hardware requirements and expected performance for their use cases.

* fix README

* docs: clean up performance table and improve formatting

* add pic for README

* refactor: simplify .gitmodules and backup legacy submodules

- Remove 7 legacy submodules from root .gitmodules (archive/third_party/*)
- Keep only 2 active submodules for kt-kernel (llama.cpp, pybind11)
- Backup complete .gitmodules to archive/.gitmodules
- Add documentation in archive/README.md for researchers who need legacy submodules

This reduces initial clone size by ~500MB and avoids downloading unused dependencies.

* refactor: move doc/ back to root directory

Keep documentation in root for easier access and maintenance.

* refactor: consolidate all images to doc/assets/

- Move kt-kernel/assets/heterogeneous_computing.png to doc/assets/
- Remove KT-SFT/assets/ (images already in doc/assets/)
- Update KT-SFT/README.md image references to ../doc/assets/
- Eliminates ~7.9MB image duplication
- Centralizes all documentation assets in one location

* fix pic path for README
2025-11-10 17:42:26 +08:00

79 lines
2.4 KiB
Python

import time
def format_time(seconds):
units = [
("hours", 3600),
("minutes", 60),
("seconds", 1),
("milliseconds", 1e-3),
("microseconds", 1e-6),
]
for unit_name, unit_value in units:
if seconds >= unit_value:
time_value = seconds / unit_value
return f"{time_value:.2f} {unit_name}"
return "0 seconds" # Handle case for 0 seconds
class Profiler:
def __init__(self):
self.timers = {}
self.counters = {}
def create_timer(self, name):
self.timers[name] = {
"start_time": None,
"elapsed_time": 0,
"running": False,
}
def start_timer(self, name):
if name not in self.timers:
raise ValueError(f"Timer '{name}' does not exist.")
if self.timers[name]["running"]:
raise ValueError(f"Timer '{name}' is already running.")
self.timers[name]["start_time"] = time.time()
self.timers[name]["running"] = True
def pause_timer(self, name):
if name not in self.timers:
raise ValueError(f"Timer '{name}' does not exist.")
if not self.timers[name]["running"]:
raise ValueError(f"Timer '{name}' is not running.")
self.timers[name]["elapsed_time"] += time.time() - self.timers[name]["start_time"]
self.timers[name]["running"] = False
def get_timer_sec(self, name):
if name not in self.timers:
raise ValueError(f"Timer '{name}' does not exist.")
if self.timers[name]["running"]:
current_time = self.timers[name]["elapsed_time"] + (time.time() - self.timers[name]["start_time"])
else:
current_time = self.timers[name]["elapsed_time"]
return current_time
def get_all_timers(self):
all_timers = {}
for name in self.timers:
all_timers[name] = self.get_timer_sec(name)
return all_timers
def report_timer_string(self, name):
return f"{name} elapsed time: {format_time(self.get_timer_sec(name))}"
def create_and_start_timer(self, name):
self.create_timer(name)
self.start_timer(name)
# Counter
def inc(self,key:str,delta:int=1):
self.counters[key] = self.counters.get(key,0) + delta
def set_counter(self,key:str,to=0):
self.counters[key] = to
def get_counter(self,key:str):
return self.counters.get(key,0)