concerns surrounding using timestamp with Put

[Douglas Campbell <de...@yahoo.com>]

Hi,

We're using 20.3 and would like to use a timestamp provided by an upstream 
system during the put for updates to record column values.

Assuming the timestamp is trusted or can be identified as invalid by some means, 
are there any other issues to be aware of when using the timestamp form of the 
put call?

The main reason we're interested in this is that it allows for out of order 
processing of updates.   i.e. an older update to a field arrives after an 
already processed one.  In this case, a "normal" read on the column value will 
produce the correct column value.

-Doug



      

RE: concerns surrounding using timestamp with Put

[Jonathan Gray <jg...@facebook.com>]

Hi Doug,

Out of order insertion of timestamps is supported in 0.89/0.90/trunk but not fully supported in the 0.20.x series.  Primarily, you can see some weird stuff using Gets in 0.20 if you do out of order timestamp insertion.  Scans are mostly okay.

JG

> -----Original Message-----
> From: Douglas Campbell [mailto:deegs_ca@yahoo.com]
> Sent: Wednesday, September 08, 2010 11:14 AM
> To: hbase users
> Subject: concerns surrounding using timestamp with Put
> 
> Hi,
> 
> We're using 20.3 and would like to use a timestamp provided by an
> upstream
> system during the put for updates to record column values.
> 
> Assuming the timestamp is trusted or can be identified as invalid by
> some means,
> are there any other issues to be aware of when using the timestamp form
> of the
> put call?
> 
> The main reason we're interested in this is that it allows for out of
> order
> processing of updates.   i.e. an older update to a field arrives after
> an
> already processed one.  In this case, a "normal" read on the column
> value will
> produce the correct column value.
> 
> -Doug
> 
> 
> 
>