The cluster node with one data node are both on the same physical machine.

On Thu, Jan 16, 2014 at 9:08 PM, Eldon Carman <[email protected]> wrote:

> Yes. VXQuery is using a cluster node with one data node. The data node is
> using a single partition.
>
>
> On Thu, Jan 16, 2014 at 8:41 PM, Vinayak Borkar <[email protected]>wrote:
>
>> I am assuming you are running VXQuery to have only 1 partition, correct?
>>
>> Vinayak
>>
>>
>>
>>
>>
>>
>> On 1/16/14, 7:14 PM, Eldon Carman wrote:
>>
>>> I was able to resolve the issue that blocked the two aggregate queries:
>>> sum
>>> (q02) and max (q03). The following queries show a similar story as the
>>> previous filter queries.
>>>
>>> Query q02 (50mb)
>>> ---------------
>>> 0m36.713s Saxon
>>> 1m16.764s VXQuery xml
>>> 1m21.594s VXQuery xml.gz
>>>
>>> Query q02 (500mb)
>>> ---------------
>>> 2m11.029s Saxon
>>> 8m03.735s VXQuery xml
>>> 8m46.822s VXQuery xml.gz
>>>
>>> Query q03 (50mb)
>>> ---------------
>>> 0m33.280s Saxon
>>> 1m00.807s VXQuery xml
>>> 1m03.500s VXQuery xml.gz
>>>
>>> Query q03 (500mb)
>>> ---------------
>>> 1m58.784s Saxon
>>> 10m01.340s VXQuery xml
>>> 10m26.236s VXQuery xml.gz
>>>
>>> Throughput:
>>> Saxon (both queries) ~1.4 mb/s (50mb) to ~3.8 mb/s (500mb)
>>> VXQuery (q02) ~0.6 mb/s
>>> VXQuery (q03) ~0.8 mb/s
>>>
>>>
>>
>

Reply via email to