Hi Mauricio, If you haven't already, take a look at the READ_AT_SNAPSHOT read mode (more info here <https://kudu.apache.org/docs/transaction_semantics.html#_read_operations_scans>). IIUC, it seems similar to, if not exactly what you're looking for!
Andrew On Mon, Nov 27, 2017 at 5:02 PM, Mauricio Aristizabal < mauri...@impactradius.com> wrote: > Hi all, has there been any talk of supporting this any time soon? > > Time travel reads are such a cool feature, but even more than in ETL jobs > (via Java/Scala), they would be most useful via SQL to ensure consistency > when reading. > > Specifically, for example our spark streaming job updates dozens of > aggregation tables every 30 seconds. To make the data fully consistent we > would love to have views over these aggs tagged with the exact timestamp we > want to expose. When each batch is done and all tables updated, we would > update all the views forward, effectively hiding the updates we're doing > until they're all ready. > > -m > > > > -- > *MAURICIO ARISTIZABAL* > Architect - Business Intelligence + Data Science > mauri...@impactradius.com(m)+1 323 309 4260 <(323)%20309-4260> > 223 E. De La Guerra St. | Santa Barbara, CA 93101 > <https://maps.google.com/?q=223+E.+De+La+Guerra+St.+%7C+Santa+Barbara,+CA+93101&entry=gmail&source=g> > > Overview <http://www.impactradius.com/?src=slsap> | Twitter > <https://twitter.com/impactradius> | Facebook > <https://www.facebook.com/pages/Impact-Radius/153376411365183> | LinkedIn > <https://www.linkedin.com/company/impact-radius-inc-> > -- Andrew Wong
