aa67a19e99
Implement new CPU detection using cpuid, cpuidex plan9 instructions from klauspost/cpuid project, remove C code
10 years ago
0a827305ad
Implement metadata cache, metadata cache is used by top level donut right now. Rename trove as data cache
...
We should use it internally everywhere.
10 years ago
14844f48dd
Make donut do everything as an atomic operation, this avoids all the deadlocks and races
10 years ago
30fc14e703
Restructure codebase move crypto, checksum to top-level, move ``split`` into donut, move crypto/keys into api/auth
10 years ago
8a4e7bcdcf
Add full API tests, move storage/donut to donut, add disk tests as well
10 years ago
7c37e9d06a
Make donut fully integrated back into API handlers
10 years ago
12bde7df30
Add simple Ticket Master which pro-actively sends messages on proceedChannel
...
Handlers are going to wait on proceedChannel, this the initial step towards
providing priority for different set of API operations
10 years ago
5cfb05465e
Add cache, donut tests separately - fix behavior differences
...
Remove priority queue, implement it using a simpler channels
10 years ago
ebe61d99d9
Use cache Append() for saving objects in memory, GetObject() caches un-cached entries while reading
10 years ago
bce93c1b3a
Integrate cache with donut, add tests
10 years ago
0533abf6a8
Make priority queue lambda function return error over a channel
10 years ago
38a6ce36e5
Remove slow AppendUniq code, rolling through over a slice is in-efficient
...
Remove it and use map instead
10 years ago
84810162f5
Add simple Version and GetSysInfo services
10 years ago
14ec42d646
Add initial implementation of priority queue, uses container/heap
10 years ago
eb5aa19dfa
Remove custom Config, will use quick Config instead for user access keys
10 years ago
701c3e5242
Add new RPC helpers wrapping over regular rpc packages, add middleware chaining ability
10 years ago
4addf7a996
Restructure API handlers, add JSON RPC simple HelloService right now.
10 years ago
335c7827eb
More donut, cache, api cleanup
10 years ago
dc0df3dc0e
Breakaway from driver model, move cache into donut
10 years ago
72572d6c71
Remove some api server code bringing in new cleanup
10 years ago
c2031ca066
Add server and control command
10 years ago
cfbc169034
Fixing API definitions
10 years ago
8f61d6b6be
Move memory code out, add it as layer on top of existing donut code
...
Just like how http.Handlers can be overlayed on top of each other
with each implementing ServeHTTP().
drivers.Driver can be overlayed on top of each other in similar manner
which would implement the drivers.Driver interface.
API <----> cache <----> donut <----> donut(format)
10 years ago
ab6e16bb41
Add dummy driver for community to submit new drivers
10 years ago
12de98fb62
Rename memory driver as cache
10 years ago
2571342451
Filesystem goes the high road *again*
10 years ago
f74d6138da
IsValidBucket() is sufficient we don't need to verify for "."
10 years ago
3109909355
Handle couple of cases of OOM conditions, move caching to GetObject() rather than PutObject()
10 years ago
d07d0c670a
Return back proper errors in writeObjectData(), rename few functions
10 years ago
10c807f233
Put object on successful write returns full metadata, to avoid subsequent GetObjectMetadata() calls in driver
10 years ago
6921328b93
Avoid frivolous GetObjectMetadata() calls at driver level, return back all the information in donut ListObjects()
10 years ago
d8f7896a43
Expand http server struct to store more values
10 years ago
42c0287943
Use errorChannels only for services not for drivers, reduce them to use simple functions
10 years ago
91e5f648cb
Move to Set() not Append() due to large memory reference copy
10 years ago
ac4d8fe478
Add free() method for proxyReader to aggressively de-allocate Read data, to handle certain out of memory conditions
...
There are still some more out there
10 years ago
05f8654e3d
Add Append() method to trove cache for appending data to an existing key
...
This largely avoids a large buffer copy which would accumulate inside proxyReader{}
This patch also implements "initialize()" function to init and populate data
on all the existing buckets, avoiding the redundant ListBuckets() invoked by
every API call.
10 years ago
367772b988
Make sure to populate on disk data into memory upon first API requests
10 years ago
45a7eab804
An attempt to bring in memory layer into donut driver
10 years ago
f3c25bcfc4
Keeping the lexical order same add optimizations, provide a comprehensive response from ListObjects()
10 years ago
ae66ae42c4
Rename functions for their purpose
10 years ago
9a4680475f
Object metadata was wrongly misconstrued to be mutable, handle it
10 years ago
3aa6d90c5e
Avoid racy maps, read from disk on success return quickly. Many more optimizations
10 years ago
68974918ac
Donut moves to typed metadata, removing the necessity for strconv, parsing and string converstions
10 years ago
767d3743ee
Add new metadata definitions and use them
10 years ago
9958e34772
Import quick key value store from Minio Client for persistent state files, primarily for donut
10 years ago
fb9adb5524
Donut cleanup another set
...
- Make sure to close all readers
- Fix errors in api_testsuite c.Assert(err, IsNil) should be done right after each function call
10 years ago
eec66f195a
Take all the ListObjects into bucket handlers
...
Earlier the listing would wait for all the objects to be processed
this is essentially very time consuming considering even for 100,000
files.
10 years ago
45e9d25931
Remove more bloated code - simplify
10 years ago
82dcbf262d
Add simple locking for donut API for now - fixes #671
10 years ago
03b4d3b308
Go vet fixes for donut
10 years ago
Harshavardhana