Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>A1. Optimize the threads per block is basically heuristics. You could try </p> <pre><code>for(int threadsPerBlock=32; threadsPerBlock&lt;=512;threadsPerBlock+=32){...} </code></pre> <p>A2. Currently you use one thread per row and sum the elements to <code>squareEuclDist</code> linearly. You could consider use one thread block per row. Within the block, each thread computes the square-difference of one element and you could use parallel reduction to sum them together. Please refer to the following link for parallel reduction.</p> <p><a href="http://docs.nvidia.com/cuda/samples/6_Advanced/reduction/doc/reduction.pdf" rel="nofollow">http://docs.nvidia.com/cuda/samples/6_Advanced/reduction/doc/reduction.pdf</a></p> <p>A3. the list you show is the total amount of global/shared memory. Multiple threads will share these hardware resources. You could find this tool in your cuda installation dir to help you calculate the exact number per thread of those hardware resources you can use in a particular kernel.</p> <pre><code>$CUDA_HOME/tools/CUDA_Occupancy_Calculator.xls </code></pre> <p>A4. <code>maximum sizes of each dimension</code> does not mean all dimensions can reach their max at the same time. However there's no limitation on block per grid, so 65536x65536x1 blocks in a grid is possible.</p> <p>A5. mem clock has nothing to do with the thread number. You could read the programming model section in cuda doc for more info.</p> <p><a href="http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#scalable-programming-model" rel="nofollow">http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#scalable-programming-model</a> </p>
    singulars
    1. This table or related slice is empty.
    1. POSome basic CUDA enquiries
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USkangshiyin
    1. USkangshiyin
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POSome basic CUDA enquiries
      singulars
      1. PTQuestion
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTAcceptedByOriginator
    1. COThe third question was a little confusing. Mainly, I was interested how the 512 MB of global memory is grouped into the shared memory , the constant memory and the other type of memories. In other words what are the vars of the equation x (Bytes) + y (Bytes) + z (Bytes) +...= 512 MB. x,y,z refers to the type of memories. Thanks.
      singulars
      1. PO
      1. USDarkmoor
    2. CO@darkmoor they are separate memories. 512M is the size of global mem. It does not include any other types of mem. Each type of mem have its own capacity.
      singulars
      1. PO
      1. USkangshiyin
    3. COCould you please give me more details about "_could consider use one thread block per row_". An code sample it would be great.
      singulars
      1. PO
      1. USDarkmoor
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload