Re: Please Help: Amplab Benchmark Performance

2014-01-29 Thread Chen Jin
a straightforward Spark RDD implementation of the same. -- Christopher T. Nguyen Co-founder CEO, Adatao linkedin.com/in/ctnguyen On Wed, Jan 29, 2014 at 8:10 PM, Chen Jin karen...@gmail.com wrote: Hi All, https://amplab.cs.berkeley.edu/benchmark/ has given a nice benchmark report. I am trying

Re: how to set SPARK_WORKER_INSTANCES and SPARK_WORKER_CORES otpimally

2014-01-26 Thread Chen Jin
? No, (a) gives u 8 workers with each core per worker (b) gives 1 worker while each worker has eight cores. Let me know, if any doubts. Thanks and Regards, Archit Thakur. On Sun, Jan 26, 2014 at 5:58 AM, Chen Jin karen...@gmail.com wrote: Hi all, From spark document, we can set the number

how to set SPARK_WORKER_INSTANCES and SPARK_WORKER_CORES otpimally

2014-01-25 Thread Chen Jin
Hi all, From spark document, we can set the number of workers by SPARK_WORKER_INSTANCES and the max number of cores that worker can take by using SPARK_WORKER_CORES, if I have 5 8-core machine, which one would perform better between (a) SPARK_WORKER_INSTANCES = 8 SPARK_WORKER_CORES = 1 and

Please help: change $SPARK_HOME/work directory for spark applications

2014-01-15 Thread Chen Jin
Hi, Currently my application jars and logs are stored in $SPARK_HOME/work, I would like to change it to somewhere having more space. Could anyone advise me on this? Changing the log dir is straightforward which just to export SPARK_LOG_DIR, however, there is no environment variable for

Re: Please help: virtualization type 'hvm' when I try to launch ec2 ssd instance

2014-01-11 Thread Chen Jin
Yeah, that's was the problem, I fixed it, thanks a lot. On Sat, Jan 11, 2014 at 4:41 PM, Patrick Wendell pwend...@gmail.com wrote: You may need to set the master to be one of the HVM types as well. What type is the master in your cluster? On Fri, Jan 10, 2014 at 2:15 PM, Chen Jin karen

Please help: virtualization type 'hvm' when I try to launch ec2 ssd instance

2014-01-10 Thread Chen Jin
Hi , I am trying to run spark with EC2 SSD instance i2.2xlarge, But, when I try to launch EC2 cluster using spark-ec2 script with --instance-type=i2.2x.large I received the following error: ResponseErrorsErrorCodeInvalidParameterCombination/CodeMessageVirtualization type 'hvm' is required for

Re: Please help: virtualization type 'hvm' when I try to launch ec2 ssd instance

2014-01-10 Thread Chen Jin
? ResponseErrorsErrorCodeInvalidParameterCombination/CodeMessageNon-Windows instances with a virtualization type of 'hvm' are currently not supported for this instance type./Message/Error/ErrorsRequestID60aa8aa0-acf3-4bc8-b2b4-120f0b7d1e36/RequestID/Response On Fri, Jan 10, 2014 at 1:40 PM, Chen Jin karen...@gmail.com wrote: Thanks, the patch