amplab/benchmark

Large scale query engine benchmark

benchmark's Issues

incorrect price per hour for m2.4xlarge

In index.md the m2.4xlarge price per hour is quoted as $0.41 but it's $1.640 (but correctly states the total cluster cost is 5 x $1.640 = $8.20)

Missing S3 data

All of the data except for the 5node sequence-snappy crawl data seems to be missing from S3. I'm accessing S3 directly using wget and my web browser, both from outside of AWS.

See the listing here:
https://big-data-benchmark.s3.amazonaws.com/

How is the dataset size measured?

When looking at the size of the Rankings table, the website/docs say it is 6.38GB, however it seems it is more in the range of 5.2GB when looking at the on disk size as well as Spark's reported size.
The format I downloaded was the txt-format, which I expected would be close to the reported 6.38GB since there is no compression.

S3 Source Data Buckets - AccessDenied

Running prepare_benchmark.sh per instructions and get "AccessDenied" on source S3 buckets and objects - specifically attempting to use the Redshift load option. Have verified same ACL problem/symptoms by using independent S3 client against s3://big-data-benchmark/pavlo/*. Note - I have used AWS IAM user keys which have full S3 read privileges. Please advise - Thanks!

Is the dataset on S3 still available nowadays?

I try to access the dataset on my aws ec2 instance (us-east-2 region) but failed.
I have configured the access key and secret key.
Here are the outputs:

# ubuntu @ ip-172-31-33-149 in ~ [14:58:32] C:77
$ aws s3 ls s3://big-data-benchmark/pavlo/text/tiny/

An error occurred (AccessDenied) when calling the ListObjectsV2 operation: Access Denied

# ubuntu @ ip-172-31-33-149 in ~ [14:58:45] C:255
$ s3cmd ls s3://big-data-benchmark/pavlo/text-deflate/tiny/
ERROR: Access to bucket 'big-data-benchmark' was denied
ERROR: S3 error: 403 (AccessDenied): Access Denied

# ubuntu @ ip-172-31-33-149 in ~ [14:59:06] C:77
$ s3cmd ls s3n://big-data-benchmark/pavlo/text-deflate/tiny/

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
    An unexpected error has occurred.
  Please try reproducing the error using
  the latest s3cmd code from the git master
  branch found at:
    https://github.com/s3tools/s3cmd
  and have a look at the known issues list:
    https://github.com/s3tools/s3cmd/wiki/Common-known-issues-and-their-solutions-(FAQ)
  If the error persists, please report the
  following lines (removing any private
  info as necessary) to:
   [email protected]


!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Invoked as: /usr/bin/s3cmd ls s3n://big-data-benchmark/pavlo/text-deflate/tiny/
Problem: <class 'ValueError: s3n://big-data-benchmark/pavlo/text-deflate/tiny/: not a recognized URI

I wander if the dataset is still available on S3 now or is there something wrong with my operation?

BDB dataset generator code

Hi I want to test my system performance with BDB query 1,2,3 over different dataset size(10k, 100k, 1million, 10million). How can I generate on my own? Where can I find the source code for generating it?

Shark is not using sequence-snappy as input

Compared to the other experiment, the Shark experiment does not use snappy formatted input.

external query - udf - not working

Hello,

I am using Hadoop 1.0 via Cloudera CDH 4.6 and am attempting to run the AMPLab benchmark on Hive.

I completed the scan, aggregate and join query sets successfully. But I am running into failures when running the external query. This is the output for the query:

ogging initialized using configuration in jar:file:/opt/cloudera/parcels/CDH-4.6.0-1.cdh4.6.0.p0.26/lib/hive/lib/hive-common-0.10.0-cdh4.6.0.jar!/hive-log4j.properties
Hive history file=/tmp/hdfs/hive_job_log_63d59f7d-5207-42a4-aee5-2baa7695cafd_81302959.txt
Total MapReduce jobs = 3
Launching Job 1 out of 3
Number of reduce tasks is set to 0 since there's no reduce operator
Starting Job = job_201407241331_0006, Tracking URL = http://hadoop1.sdcorp.global.sandisk.com:50030/jobdetails.jsp?jobid=job_201407241331_0006
Kill Command = /opt/cloudera/parcels/CDH-4.6.0-1.cdh4.6.0.p0.26/lib/hadoop/bin/hadoop job -kill job_201407241331_0006
Hadoop job information for Stage-1: number of mappers: 498; number of reducers: 0
2014-07-25 10:32:14,277 Stage-1 map = 0%, reduce = 0%
2014-07-25 10:33:01,617 Stage-1 map = 100%, reduce = 100%
Ended Job = job_201407241331_0006 with errors
Error during job, obtaining debugging information...
Job Tracking URL: http://hadoop1.sdcorp.global.sandisk.com:50030/jobdetails.jsp?jobid=job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000499 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000039 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000032 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000006 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000016 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000062 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000046 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000061 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000018 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000025 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000064 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000033 (and more) from job job_201407241331_0006

Examining task ID: task_201407241331_0006_m_000059 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000245 (and more) from job job_201407241331_0006
Examining task ID: task_201407241331_0006_m_000056 (and more) from job job_201407241331_0006

Task with the most failures(4):

Task ID:
task_201407241331_0006_m_000032

URL:

http://hadoop1.sdcorp.global.sandisk.com:50030/taskdetails.jsp?jobid=job_201407241331_0006&tipid=task_201407241331_0006_m_000032

Diagnostic Messages for this Task:

FAILED: Execution Error, return code 2 from org.apache.hadoop.hive.ql.exec.MapRedTask
MapReduce Jobs Launched:
Job 0: Map: 498 HDFS Read: 0 HDFS Write: 0 FAIL
Total MapReduce CPU Time Spent: 0 msec

Am I missing some setup for running the external query? Can someone help?

Thanks,
Madhura

No Licence

Hi,

my Name is Dimitri Vorona and I am from the Technische Universität in Munich.

We'd like to reuse your deployment code (with appropriate credits, of course), yet, can not do it for the apparent lack of licencing.

If you don't have any objections for the code to be reused, I'd be very thankful, if you'd explicitly specify a licence somewhere.