You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@jena.apache.org by Alan Wu <al...@oracle.com> on 2014/03/20 23:22:00 UTC
A question on concurrent access to Fuseki
Hi,
Please kindly respond to my email address as I am not part of the
mailing list.
When concurrent requests are sent to the same Fuseki (1.0.1) endpoint,
will there be multiple Dataset objects serving those concurrent requests?
Thanks,
Zhe
Re: A question on concurrent access to Fuseki
Posted by Alan Wu <al...@oracle.com>.
Hi Andy,
Thanks. It helps,
Zhe
On 3/21/2014 2:02 AM, Andy Seaborne wrote:
> On 20/03/14 22:22, Alan Wu wrote:
>> Hi,
>>
>> Please kindly respond to my email address as I am not part of the
>> mailing list.
>
> Zhe - please subscribe to see any further replies. I've copied you on
> this one.
>
>>
>> When concurrent requests are sent to the same Fuseki (1.0.1) endpoint,
>> will there be multiple Dataset objects serving those concurrent
>> requests?
>
> No.
>
> The DatasetGraph is shared across all requests to the same data
> service, concurrent or not.
>
> In Fuseki1, it is created once at startup.
>
> In Fuseki2, it is created once when a data service starts. This may
> be at startup or sometime later if the data service is marked inactive
> and then activated.
>
> DatasetGraph, and the associated machinery around them, e.g. query
> engines, are expected to handle concurrency and transactions. (This is
> true everywhere - not just Fuseki.)
>
> Andy
>
>>
>> Thanks,
>>
>> Zhe
>
Re: A question on concurrent access to Fuseki
Posted by Andy Seaborne <an...@apache.org>.
On 20/03/14 22:22, Alan Wu wrote:
> Hi,
>
> Please kindly respond to my email address as I am not part of the
> mailing list.
Zhe - please subscribe to see any further replies. I've copied you on
this one.
>
> When concurrent requests are sent to the same Fuseki (1.0.1) endpoint,
> will there be multiple Dataset objects serving those concurrent requests?
No.
The DatasetGraph is shared across all requests to the same data service,
concurrent or not.
In Fuseki1, it is created once at startup.
In Fuseki2, it is created once when a data service starts. This may be
at startup or sometime later if the data service is marked inactive and
then activated.
DatasetGraph, and the associated machinery around them, e.g. query
engines, are expected to handle concurrency and transactions. (This is
true everywhere - not just Fuseki.)
Andy
>
> Thanks,
>
> Zhe