User Tools

Site Tools


cluster:174

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
cluster:174 [2018/08/27 10:33]
hmeij07
cluster:174 [2018/08/28 08:38] (current)
hmeij07
Line 3: Line 3:
  
 ==== SQL on GPU ==== ==== SQL on GPU ====
 +
 +MapD built the first ever open source SQL engine to harness GPU computing for analytics. Designed for maximum performance,​ the MapD SQL engine dynamically compiles SQL to run across multiple GPUs and CPUs. Massively parallel database servers.
 +
 +  * http://​www.mapd.com
 +
 +** Browser **
 +
 +Log into one of the tails like petaltail, swallowtail or cottontail2 preferably, and with x11-forwarding enabled go to port 9092 of node ''​n37''​ and create a dashboard. Use the "​data"​ subnet (10.10) eth1.
 +
 +  * http://​n37-eth1:​9092
 +
 +  - Click New Dashboard.
 +  - Click Add Chart.
 +  - Click SCATTER.
 +  - Click Add Data Source.
 +  - Choose the flights_2008_7M table as the data source.
 +  - Click X Axis +Add Measure.
 +  - Choose depdelay. (departure delay)
 +  - Click Y Axis +Add Measure.
 +  - Choose arrdelay. (arrival delay)
 +
 +7 million records, amazingly fast. I mean  like < 50 ms (milliseconds). There are pie, bar, stacked bar, bubble etc charts and point maps or chloropleth maps. A SQL editor. ​ And I do not even observe a blip on gpu utilization. Each of these K20 nodes still has 7 drive bays available which would allow for a large database storage platform if we had a mapd project.
 +
 +{{ :​cluster:​dep-arr-scatterplot.png?​nolink&​400 |}}
 +
  
 ** Interactive ** ** Interactive **
  
-https://​www.mapd.com/​docs/​latest/​3_mapdql.html+  * https://​www.mapd.com/​docs/​latest/​3_mapdql.html 
 +  * can be scripted using mapdql 
 + 
 +<​code>​ 
 + 
 +[root@n37 ~]# export MAPD_USER=mapd 
 +[root@n37 ~]# export MAPD_GROUP=mapd 
 +[root@n37 ~]# export MAPD_STORAGE=/​var/​lib/​mapd 
 +[root@n37 ~]# export MAPD_PATH=/​opt/​mapd 
 +[root@n37 ~]# cd $MAPD_PATH 
 +[root@n37 mapd]# ./​bin/​mapdql 
 +Password: (HyperInteractive) 
 +User mapd connected to database mapd 
 +mapdql>
  
 mapdql> \t ^flight.* mapdql> \t ^flight.*
 flights_2008_7M flights_2008_7M
  
-mapdql> SELECT origin_city AS "​Origin",​ dest_city AS "​Destination",​ AVG(airtime) AS "​Average Airtime"​ FROM flights_2008_7M WHERE distance < 175 GROUP BY origin_city,​ dest_city;+mapdql> SELECT origin_city AS "​Origin",​ dest_city AS "​Destination", ​ 
 +AVG(airtime) AS "​Average Airtime"​ FROM flights_2008_7M ​ 
 +WHERE distance < 175 GROUP BY origin_city,​ dest_city;
 ... ...
-SELECT origin_city AS "​Origin",​ dest_city AS "​Destination",​ AVG(airtime) AS +Portland|North Bend|46.12162162162162 
-"​Average Airtime"​ FROM flights_2008_10k WHERE distance < 175 GROUP BY origin_city,​ +Medford|North Bend|28 
-dest_city; +Covington|Huntington|24.98076923076923 
 +mapdql> ​
  
 +</​code>​
  
  
 ** Demos ** ** Demos **
  
-https://​www.mapd.com/​demos/​ +  * https://​www.mapd.com/​demos/​
- +
  
 ** Load sample ** ** Load sample **
  
 <​code>​ <​code>​
-/opt/mapd+ 
 +/opt/mapd/​insert_sample_data 
 Enter dataset number to download, or '​q'​ to quit: Enter dataset number to download, or '​q'​ to quit:
  # ​    ​Dataset ​                  ​Rows ​   Table Name             File Name  # ​    ​Dataset ​                  ​Rows ​   Table Name             File Name
Line 58: Line 99:
 </​code>​ </​code>​
  
 +  * https://​www.mapd.com/​docs/​latest/​4_centos7-yum-gpu-ce-recipe.html
  
 \\ \\
 **[[cluster:​0|Back]]** **[[cluster:​0|Back]]**
cluster/174.1535380434.txt.gz · Last modified: 2018/08/27 10:33 by hmeij07