Source code
Revision control
Copy as Markdown
Other Tools
Metadata-Version: 2.1
Name: diskcache
Version: 4.1.0
Summary: Disk Cache -- Disk and file backed persistent cache.
Author: Grant Jenks
Author-email: contact@grantjenks.com
License: Apache 2.0
Platform: UNKNOWN
Classifier: Development Status :: 5 - Production/Stable
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Natural Language :: English
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 2
Classifier: Programming Language :: Python :: 2.7
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.4
Classifier: Programming Language :: Python :: 3.5
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: Implementation :: CPython
Classifier: Programming Language :: Python :: Implementation :: PyPy
DiskCache: Disk Backed Cache
============================
`DiskCache`_ is an Apache2 licensed disk and file backed cache library, written
in pure-Python, and compatible with Django.
The cloud-based computing of 2019 puts a premium on memory. Gigabytes of empty
space is left on disks as processes vie for memory. Among these processes is
Memcached (and sometimes Redis) which is used as a cache. Wouldn't it be nice
to leverage empty disk space for caching?
Django is Python's most popular web framework and ships with several caching
backends. Unfortunately the file-based cache in Django is essentially
broken. The culling method is random and large caches repeatedly scan a cache
directory which slows linearly with growth. Can you really allow it to take
sixty milliseconds to store a key in a cache with a thousand items?
In Python, we can do better. And we can do it in pure-Python!
::
In [1]: import pylibmc
In [2]: client = pylibmc.Client(['127.0.0.1'], binary=True)
In [3]: client[b'key'] = b'value'
In [4]: %timeit client[b'key']
10000 loops, best of 3: 25.4 µs per loop
In [5]: import diskcache as dc
In [6]: cache = dc.Cache('tmp')
In [7]: cache[b'key'] = b'value'
In [8]: %timeit cache[b'key']
100000 loops, best of 3: 11.8 µs per loop
**Note:** Micro-benchmarks have their place but are not a substitute for real
measurements. DiskCache offers cache benchmarks to defend its performance
claims. Micro-optimizations are avoided but your mileage may vary.
DiskCache efficiently makes gigabytes of storage space available for
caching. By leveraging rock-solid database libraries and memory-mapped files,
cache performance can match and exceed industry-standard solutions. There's no
need for a C compiler or running another process. Performance is a feature and
testing has 100% coverage with unit tests and hours of stress.
Testimonials
------------
`Daren Hasenkamp`_, Founder --
"It's a useful, simple API, just like I love about Redis. It has reduced
the amount of queries hitting my Elasticsearch cluster by over 25% for a
website that gets over a million users/day (100+ hits/second)."
`Mathias Petermann`_, Senior Linux System Engineer --
"I implemented it into a wrapper for our Ansible lookup modules and we were
able to speed up some Ansible runs by almost 3 times. DiskCache is saving
us a ton of time."
Does your company or website use `DiskCache`_? Send us a `message
<contact@grantjenks.com>`_ and let us know.
Features
--------
- Pure-Python
- Fully Documented
- Benchmark comparisons (alternatives, Django cache backends)
- 100% test coverage
- Hours of stress testing
- Performance matters
- Django compatible API
- Thread-safe and process-safe
- Supports multiple eviction policies (LRU and LFU included)
- Keys support "tag" metadata and eviction
- Developed on Python 3.7
- Tested on CPython 2.7, 3.4, 3.5, 3.6, 3.7 and PyPy
- Tested on Linux, Mac OS X, and Windows
- Tested using Travis CI and AppVeyor CI
Quickstart
----------
$ pip install diskcache
You can access documentation in the interpreter with Python's built-in help
function::
>>> import diskcache
>>> help(diskcache)
The core of `DiskCache`_ is three data types intended for caching. `Cache`_
objects manage a SQLite database and filesystem directory to store key and
value pairs. `FanoutCache`_ provides a sharding layer to utilize multiple
caches and `DjangoCache`_ integrates that with `Django`_::
>>> from diskcache import Cache, FanoutCache, DjangoCache
>>> help(Cache)
>>> help(FanoutCache)
>>> help(DjangoCache)
Built atop the caching data types, are `Deque`_ and `Index`_ which work as a
cross-process, persistent replacements for Python's ``collections.deque`` and
``dict``. These implement the sequence and mapping container base classes::
>>> from diskcache import Deque, Index
>>> help(Deque)
>>> help(Index)
Finally, a number of `recipes`_ for cross-process synchronization are provided
using an underlying cache. Features like memoization with cache stampede
prevention, cross-process locking, and cross-process throttling are available::
>>> from diskcache import memoize_stampede, Lock, throttle
>>> help(memoize_stampede)
>>> help(Lock)
>>> help(throttle)
Python's docstrings are a quick way to get started but not intended as a
replacement for the `DiskCache Tutorial`_ and `DiskCache API Reference`_.
User Guide
----------
For those wanting more details, this part of the documentation describes
tutorial, benchmarks, API, and development.
* `DiskCache Tutorial`_
* `DiskCache Cache Benchmarks`_
* `DiskCache DjangoCache Benchmarks`_
* `Case Study: Web Crawler`_
* `Case Study: Landing Page Caching`_
* `Talk: All Things Cached - SF Python 2017 Meetup`_
* `DiskCache API Reference`_
* `DiskCache Development`_
.. _`DiskCache DjangoCache Benchmarks`: http://www.grantjenks.com/docs/diskcache/djangocache-benchmarks.html
.. _`Talk: All Things Cached - SF Python 2017 Meetup`: http://www.grantjenks.com/docs/diskcache/sf-python-2017-meetup-talk.html
.. _`Case Study: Landing Page Caching`: http://www.grantjenks.com/docs/diskcache/case-study-landing-page-caching.html
Comparisons
-----------
Comparisons to popular projects related to `DiskCache`_.
Key-Value Stores
................
`DiskCache`_ is mostly a simple key-value store. Feature comparisons with four
other projects are shown in the tables below.
* `dbm`_ is part of Python's standard library and implements a generic
interface to variants of the DBM database — dbm.gnu or dbm.ndbm. If none of
these modules is installed, the slow-but-simple dbm.dumb is used.
* `shelve`_ is part of Python's standard library and implements a “shelf” as a
persistent, dictionary-like object. The difference with “dbm” databases is
that the values can be anything that the pickle module can handle.
* `sqlitedict`_ is a lightweight wrapper around Python's sqlite3 database with
a simple, Pythonic dict-like interface and support for multi-thread
access. Keys are arbitrary strings, values arbitrary pickle-able objects.
* `pickleDB`_ is a lightweight and simple key-value store. It is built upon
Python's simplejson module and was inspired by Redis. It is licensed with the
BSD three-caluse license.
**Features**
================ ============= ========= ========= ============ ============
Feature diskcache dbm shelve sqlitedict pickleDB
================ ============= ========= ========= ============ ============
Atomic? Always Maybe Maybe Maybe No
Persistent? Yes Yes Yes Yes Yes
Thread-safe? Yes No No Yes No
Process-safe? Yes No No Maybe No
Backend? SQLite DBM DBM SQLite File
Serialization? Customizable None Pickle Customizable JSON
Data Types? Mapping/Deque Mapping Mapping Mapping Mapping
Ordering? Insert/Sorted None None None None
Eviction? LRU/LFU/more None None None None
Vacuum? Automatic Maybe Maybe Manual Automatic
Transactions? Yes No No Maybe No
Multiprocessing? Yes No No No No
Forkable? Yes No No No No
Metadata? Yes No No No No
================ ============= ========= ========= ============ ============
**Quality**
================ ============= ========= ========= ============ ============
Project diskcache dbm shelve sqlitedict pickleDB
================ ============= ========= ========= ============ ============
Tests? Yes Yes Yes Yes Yes
Coverage? Yes Yes Yes Yes No
Stress? Yes No No No No
CI Tests? Linux/Windows Yes Yes Linux No
Python? 2/3/PyPy All All 2/3 2/3
License? Apache2 Python Python Apache2 3-Clause BSD
Docs? Extensive Summary Summary Readme Summary
Benchmarks? Yes No No No No
Sources? GitHub GitHub GitHub GitHub GitHub
Pure-Python? Yes Yes Yes Yes Yes
Server? No No No No No
Integrations? Django None None None None
================ ============= ========= ========= ============ ============
**Timings**
These are rough measurements. See `DiskCache Cache Benchmarks`_ for more
rigorous data.
================ ============= ========= ========= ============ ============
Project diskcache dbm shelve sqlitedict pickleDB
================ ============= ========= ========= ============ ============
get 25 µs 36 µs 41 µs 513 µs 92 µs
set 198 µs 900 µs 928 µs 697 µs 1,020 µs
delete 248 µs 740 µs 702 µs 1,717 µs 1,020 µs
================ ============= ========= ========= ============ ============
Caching Libraries
.................
* `joblib.Memory`_ provides caching functions and works by explicitly saving
the inputs and outputs to files. It is designed to work with non-hashable and
potentially large input and output data types such as numpy arrays.
* `klepto`_ extends Python’s `lru_cache` to utilize different keymaps and
alternate caching algorithms, such as `lfu_cache` and `mru_cache`. Klepto
uses a simple dictionary-sytle interface for all caches and archives.
Data Structures
...............
* `dict`_ is a mapping object that maps hashable keys to arbitrary
values. Mappings are mutable objects. There is currently only one standard
Python mapping type, the dictionary.
* `pandas`_ is a Python package providing fast, flexible, and expressive data
structures designed to make working with “relational” or “labeled” data both
easy and intuitive.
* `Sorted Containers`_ is an Apache2 licensed sorted collections library,
written in pure-Python, and fast as C-extensions. Sorted Containers
implements sorted list, sorted dictionary, and sorted set data types.
Pure-Python Databases
.....................
* `ZODB`_ supports an isomorphic interface for database operations which means
there's little impact on your code to make objects persistent and there's no
database mapper that partially hides the datbase.
* `CodernityDB`_ is an open source, pure-Python, multi-platform, schema-less,
NoSQL database and includes an HTTP server version, and a Python client
library that aims to be 100% compatible with the embedded version.
* `TinyDB`_ is a tiny, document oriented database optimized for your
happiness. If you need a simple database with a clean API that just works
without lots of configuration, TinyDB might be the right choice for you.
Object Relational Mappings (ORM)
................................
* `Django ORM`_ provides models that are the single, definitive source of
information about data and contains the essential fields and behaviors of the
stored data. Generally, each model maps to a single SQL database table.
* `SQLAlchemy`_ is the Python SQL toolkit and Object Relational Mapper that
gives application developers the full power and flexibility of SQL. It
provides a full suite of well known enterprise-level persistence patterns.
* `Peewee`_ is a simple and small ORM. It has few (but expressive) concepts,
making it easy to learn and intuitive to use. Peewee supports Sqlite, MySQL,
and PostgreSQL with tons of extensions.
* `SQLObject`_ is a popular Object Relational Manager for providing an object
interface to your database, with tables as classes, rows as instances, and
columns as attributes.
* `Pony ORM`_ is a Python ORM with beautiful query syntax. Use Python syntax
for interacting with the database. Pony translates such queries into SQL and
executes them in the database in the most efficient way.
SQL Databases
.............
* `SQLite`_ is part of Python's standard library and provides a lightweight
disk-based database that doesn’t require a separate server process and allows
accessing the database using a nonstandard variant of the SQL query language.
* `MySQL`_ is one of the world’s most popular open source databases and has
become a leading database choice for web-based applications. MySQL includes a
standardized database driver for Python platforms and development.
* `PostgreSQL`_ is a powerful, open source object-relational database system
with over 30 years of active development. Psycopg is the most popular
PostgreSQL adapter for the Python programming language.
* `Oracle DB`_ is a relational database management system (RDBMS) from the
Oracle Corporation. Originally developed in 1977, Oracle DB is one of the
most trusted and widely used enterprise relational database engines.
* `Microsoft SQL Server`_ is a relational database management system developed
by Microsoft. As a database server, it stores and retrieves data as requested
by other software applications.
Other Databases
...............
* `Memcached`_ is free and open source, high-performance, distributed memory
object caching system, generic in nature, but intended for use in speeding up
dynamic web applications by alleviating database load.
* `Redis`_ is an open source, in-memory data structure store, used as a
database, cache and message broker. It supports data structures such as
strings, hashes, lists, sets, sorted sets with range queries, and more.
* `MongoDB`_ is a cross-platform document-oriented database program. Classified
as a NoSQL database program, MongoDB uses JSON-like documents with
schema. PyMongo is the recommended way to work with MongoDB from Python.
* `LMDB`_ is a lightning-fast, memory-mapped database. With memory-mapped
files, it has the read performance of a pure in-memory database while
retaining the persistence of standard disk-based databases.
* `BerkeleyDB`_ is a software library intended to provide a high-performance
embedded database for key/value data. Berkeley DB is a programmatic toolkit
that provides built-in database support for desktop and server applications.
* `LevelDB`_ is a fast key-value storage library written at Google that
provides an ordered mapping from string keys to string values. Data is stored
sorted by key and users can provide a custom comparison function.
Reference
---------
* `DiskCache Documentation`_
* `DiskCache at PyPI`_
* `DiskCache at GitHub`_
* `DiskCache Issue Tracker`_
License
-------
Copyright 2016-2019 Grant Jenks
Licensed under the Apache License, Version 2.0 (the "License"); you may not use
this file except in compliance with the License. You may obtain a copy of the
License at
Unless required by applicable law or agreed to in writing, software distributed
under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR
CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.