Histogramming as a Service (HistServ)

Install from PyPI

pip install histserv

Install from conda-forge

conda install -c conda-forge histserv

Quickstart

Start the async gRPC server (or just ./example/start_server.sh):

histserv --port 50051
# 2026-03-26 11:58:00.643 INFO:histserv:server (listening at [::]:50051) started with port=50051, prune_after=24.00 h, prune_interval=5.00 min, stats_interval=5.00 s

Then run:

from hist import Hist
from histserv import Client
import numpy as np


# initialize hist locally
H_local = Hist.new.Reg(30, -3, 3, name="x", label="x-axis").Double()

with Client(address="[::]:50051") as client:
    # initialize it on the server and receive a remote hist to interact with it
    H_remote = client.init(H_local)
    # fill the remote hist; the client pre-bins locally and sends dense storage
    H_remote.fill(x=np.random.normal(size=1000))
    # retrieve it back as a ChunkedHist, drop it from the server, and materialize it locally
    H_snapshot = H_remote.snapshot(delete_from_server=True)
    print(H_snapshot.to_hist())


# local hist hasn't been filled
assert np.all(H_local.view(True) == 0)

Output in ipython:

┌────────────────────────────────────────────────────────────────────────────┐
[-inf,   -3) 1  │▋                                                           │
[  -3, -2.8) 0  │                                                            │
[-2.8, -2.6) 1  │▋                                                           │
[-2.6, -2.4) 1  │▋                                                           │
[-2.4, -2.2) 6  │████                                                        │
[-2.2,   -2) 11 │███████▍                                                    │
[  -2, -1.8) 12 │████████                                                    │
[-1.8, -1.6) 20 │█████████████▍                                              │
[-1.6, -1.4) 19 │████████████▊                                               │
[-1.4, -1.2) 33 │██████████████████████▏                                     │
[-1.2,   -1) 50 │█████████████████████████████████▌                          │
[  -1, -0.8) 70 │██████████████████████████████████████████████▉             │
[-0.8, -0.6) 49 │████████████████████████████████▉                           │
[-0.6, -0.4) 88 │███████████████████████████████████████████████████████████ │
[-0.4, -0.2) 63 │██████████████████████████████████████████▎                 │
[-0.2,    0) 65 │███████████████████████████████████████████▋                │
[   0,  0.2) 85 │█████████████████████████████████████████████████████████   │
[ 0.2,  0.4) 77 │███████████████████████████████████████████████████▋        │
[ 0.4,  0.6) 65 │███████████████████████████████████████████▋                │
[ 0.6,  0.8) 61 │████████████████████████████████████████▉                   │
[ 0.8,    1) 63 │██████████████████████████████████████████▎                 │
[   1,  1.2) 45 │██████████████████████████████▏                             │
[ 1.2,  1.4) 36 │████████████████████████▏                                   │
[ 1.4,  1.6) 32 │█████████████████████▌                                      │
[ 1.6,  1.8) 15 │██████████                                                  │
[ 1.8,    2) 11 │███████▍                                                    │
[   2,  2.2) 10 │██████▊                                                     │
[ 2.2,  2.4) 3  │██                                                          │
[ 2.4,  2.6) 5  │███▍                                                        │
[ 2.6,  2.8) 2  │█▍                                                          │
[ 2.8,    3) 1  │▋                                                           │
[   3,  inf) 0  │                                                            │
└────────────────────────────────────────────────────────────────────────────┘

Dashboard

See more information about setting up the histserv dashboard in the corresponding dashboard README.

Examples

See example/ for more examples.

Run example client:

python example/client.py
# Remote hist initialized: RemoteHist(hist_id='52c77c93da8146f2a72c53af269d1ab5', address='[::]:50051', token=None)
# Remote hist fill succeeded.
# Snapshotting current hist: ChunkedHist(...)
# Remote hist fill succeeded.
# Remote hist fill succeeded.
# Remote hist flushed successfully to hist.h5.

Or check out how to use remote histogram filling with an example coffea Processor in example/coffea_processor.py.

Useful client methods on RemoteHist:

• fill(...)
• fill_many([...])
• snapshot(delete_from_server=False)
• reset()
• exists()
• get_connection_info()
• RemoteHist.from_connection_info(...)
• flush(destination="hist.h5")
• delete()

Current supported types

Axis support:

• hist.axis.Regular without transforms
• hist.axis.Boolean
• hist.axis.Variable
• hist.axis.Integer
• hist.axis.IntCategory
• hist.axis.StrCategory

Storage support:

• hist.storage.Double
• hist.storage.Int64
• boost_histogram.storage.AtomicInt64
• hist.storage.Weight
• boost_histogram.storage.Unlimited

Unsupported today:

• transformed hist.axis.Regular
• boost_histogram.storage.Mean
• boost_histogram.storage.WeightedMean
• overflow bins on IntCategory/StrCategory chunk axes are not preserved across from_hist()/to_hist() (categorical overflow is silently dropped)

Notes:

• Growable categorical axes (IntCategory, StrCategory) are treated as chunk keys rather than dense axes.
• Histograms with one or more growable categorical axes are supported; the categorical values must be provided as scalars when filling.
• On the wire, fills are sent as dense per-chunk payloads rather than as generic Python objects.
• RemoteHist.snapshot() returns a ChunkedHist; call .to_hist() to materialize a local hist.Hist.
• fill_many(...) is useful for bundling several fills into one gRPC request.
• Dense ndarray transport is generic over NumPy dtypes, but object arrays are not supported on the wire.

Developer Info

Install

uv sync --dev

Test

python -m pytest -q
uvx ty check src

Protobuf Codegen

python -m grpc_tools.protoc -Isrc/histserv/protos --python_out=src/histserv/protos --pyi_out=src/histserv/protos --grpc_python_out=src/histserv/protos src/histserv/protos/hist.proto

After regeneration, ensure src/histserv/protos/hist_pb2_grpc.py keeps the package-relative import:

from . import hist_pb2 as hist__pb2