andi/perfdb
1
0
Fork 0
Code Issues 1 Pull Requests Packages Projects Releases Activity
458 Commits 2 Branches 0 Tags
d91481809e321b473f250fecef87d0506e9f20dd
Commit Graph

458 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
andi 4f61d91c79 draw tic for max value on y-axis only if it makes sense 2019-08-25 15:40:20 +02:00
andi a905c608aa increase maximal allowed duration for 'parallel requests' plot 2019-08-25 10:47:23 +02:00
andi da97a13ed8 add new ranges for date picker 2019-08-25 10:40:05 +02:00
andi 2cd0d71dde fix compile error in Eclipse 
Eclipse marked this as an error. A Gradle build had
no issue.
 2019-08-25 10:26:39 +02:00
andi 5b57417f75 make the tics on the y-axis easier readable 
People are having trouble to understand durations like
100000 or 2.7E+6 milliseconds. Therefore we are
hanging the labels on the y-axis to include the unit
in the tic's label. We also use multiples of seconds,
minutes, hours and days instead of multiples of 10.
 2019-08-25 10:25:47 +02:00
andi 4e9b556ea9 update third-party libs 2019-08-24 19:59:06 +02:00
andi 3a7688d1ae remember next eviction time and skip eviction 2019-08-24 19:39:59 +02:00
andi 6eaf4e10fc add maxSize parameter to HotEntryCache 2019-08-24 19:24:20 +02:00
andi 00c20dae6b use long instead of Instant for time 
Working with longs is faster and requires less
cache. The space in L123 caches is precious.
 2019-08-19 18:58:24 +02:00
andi feda901f6d remove event types 
We only have removal events. The additional complexity
of having a generic interface for many different event
types does not pay off.
 2019-08-18 20:30:25 +02:00
andi 4d9ea6d2a8 switch back to my own HotEntryCache implementation 
Guava's cache does not evict elements reliably by
time. Configure a cache to have a lifetime of n
seconds, then you cannot expect that an element is
actually evicted after n seconds with Guava.
 2019-08-18 20:14:14 +02:00
andi 0dc908c79c show the removal listener of HotEntryCache is not called on expire 2019-08-18 09:27:18 +02:00
andi 3252fcf42d improve trace logging 
- Add filename for trace logs for read/write operations.
 2019-08-18 09:25:49 +02:00
andi 16bf0933e7 update to spring-boot 2.1.7 2019-08-12 08:36:09 +02:00
andi 0b3eb97b96 Fix to string for maps with values of type Empty 
The MAX_KEY inserted into the tree had a value of one byte. This
triggered an assertion for maps with values of type Empty, because they
expected values to be empty.
Fixed by using an empty array for the value of the MAX_KEY.
 2019-08-12 08:35:40 +02:00
andi 427d7818bf update gradle and commons-collections 2019-08-01 18:54:52 +02:00
andi 574a9c9cc4 update jquery to 3.4.1 2019-07-04 19:12:54 +02:00
andi a8d77562c9 update daterangepicker to 3.0.5 2019-07-04 19:11:00 +02:00
andi 2cb81e5acd make it possible to ignore columns using the csv ingestor 2019-07-04 09:51:33 +02:00
andi 3a39f66e22 update 3rd party libs 2019-07-04 09:06:06 +02:00
andi 58f988f378 update gradle to 5.5 2019-07-04 09:01:39 +02:00
andi 4f57a29c3b increase cache of stringified longs 
This should improve the csv generation a little bit.
 2019-05-25 17:55:14 +02:00
andi 1be57dce0c update spring and jackson 2019-05-25 17:53:54 +02:00
andi 2eb2a69c17 rename 'percentile' plots to 'cumulative distribution' 2019-05-12 14:30:16 +02:00
andi 1f144846db update 3rd party libs and gradle 2019-04-28 14:40:43 +02:00
andi 9fb1a136c8 cache last used date prefix 
The 99.9999% use case is to ingest data
from the same month.
 2019-04-22 09:51:44 +02:00
andi dfe9579726 use DateTimeRange.max() instead of arbitrary relative range 2019-04-20 20:36:26 +02:00
andi d82b33c60e update js libraries 2019-04-20 20:31:51 +02:00
andi 7277670b8b update 3rd party libs 2019-04-20 20:19:24 +02:00
andi 9525ee22a0 add access restrictions for a few unwelcome classes 2019-04-20 20:12:45 +02:00
andi 56085061ed do not return anything if the field/value does not exist 
The computation of proposals is done by searching for values in a
combined index. If one of the values didn't exist, then the algorithm
returned all values. Fixed by checking that we query only existing
field/values from the combined index.
 2019-04-20 19:48:51 +02:00
andi dbe0e02517 rename cluster to partition 
We are not clustering the indices, we
are partitioning them.
 2019-04-14 10:10:16 +02:00
andi 2a1885a77f cluster the indices 2019-03-31 09:01:55 +02:00
andi 95f2f26966 handle IOExceptions earlier 2019-03-17 11:13:46 +01:00
andi 5d0ceb112e add clustering for DiskStore 2019-03-17 10:53:02 +01:00
andi b5e2d0a217 introduce clustering for query completion indices 2019-03-16 10:19:28 +01:00
andi fb9f8592ac make ClusteredPersistentMap easier to use 2019-02-24 19:20:44 +01:00
andi 59aea1a15f introduce index clustering (part 1) 
In order to prevent files from getting too big and
make it easier to implement retention policies, we
are splitting all files into chunks. Each chunk
contains the data for a time interval (1 month per
default).
This first changeset introduces the ClusteredPersistentMap
that implements this for PersistentMap. It is used
for a couple (not all) of indices.
 2019-02-24 16:50:57 +01:00
andi 372a073b6d PdbWriter is no longer in the API of DataStore 2019-02-16 16:24:14 +01:00
andi 92a47d9b56 remove TagsToFile 
Remove one layer of abstraction by moving the code into the DataStore.
 2019-02-16 16:06:46 +01:00
andi 117ef4ea34 use guava's cache as implementation for the HotEntryCache 
My own implementation was faster, but was not able to
implement a size limitation.
 2019-02-16 10:23:52 +01:00
andi 7b00eede86 refactoring: extract EncoderDecoders from DataStore 2019-02-16 09:16:15 +01:00
andi cbcb7714bb split BSFile into a TimeSeries and a LongStream file 
BSFile was used to store two types of data. This makes
the API complex. I split the API into two files with
easier and more clear APIs. Interestingly the API of
BSFile is still rather complex and has to consider both
use cases.
 2019-02-10 09:59:16 +01:00
andi fd55ea0866 update vuejs to 2.6.4 
Added the version to moment.min.js
 2019-02-09 15:39:28 +01:00
andi 93dea402a5 remove obsolete class 2019-02-09 15:25:39 +01:00
andi 27b83234cc group proposal as if they were hierarchical 
We interpret dots ('.') as hierarchy delimiter in.
That way we can reduce the number of proposed values
and show only those for the next level.
 2019-02-09 15:21:35 +01:00
andi 493971bcf3 values used in queries were added to the keys.csv 
Due to a mistake in Tag which added all strings used
by Tag into the String dictionary, the dictionary
did contain all values that were used in queries.
 2019-02-09 08:28:23 +01:00
andi ea5884a5e6 move creation of PdbWriter to the DataStore 2019-02-07 18:06:41 +01:00
andi 58bfba23bb reset lastEpochMilli when opening a new export file 2019-02-06 15:52:37 +00:00
andi 99cdf557b3 add metric logger for query completion evaluation 2019-02-06 15:51:41 +00:00