hi Samuel, Apologies for the delay in response as well as overlooking Presto mention in your initial post itself. #IMHO : Presto is lightweight, easy to install and configure. It does not support "UPDATE" .. hmm , i don't need updates in Big Data analytics where i can have a temp / intermediate table which will be faster as well (between, I don't know how many others provide true Update capabilities) I am happy with Hive itself , and don't need Presto for my ad-hoc analytics since the overhead of MR job kick-off timing is not overwhelming compared to the total query execution time. Presto is good for me when I need to run parameterized and fixed queries from a dashboard directly on my HDP cluster as it reduces my screen staring time
Hope you find it helpful in your decision making. regards Devopam On Tue, Feb 3, 2015 at 1:57 PM, Samuel Marks <samuelma...@gmail.com> wrote: > Thanks Devopam, > > In my initial post I did mention Presto, with his review: > " can query Hive, Cassandra <http://cassandra.apache.org/>, relational > DBs &etc. Doesn't seem to be designed for low-latency responses across > small clusters, or support UPDATE operations. It is optimized for data > warehousing or analytics¹ > <http://prestodb.io/docs/current/overview/use-cases.html>" > > Your thoughts? > > Best, > > Samuel Marks > http://linkedin.com/in/samuelmarks > On 03/02/2015 6:06 pm, "Devopam Mittra" <devo...@gmail.com> wrote: > >> hi Samuel, >> You may wish to evaluate Presto (https://prestodb.io/) , which has an >> added advantage of being faster than conventional Hive due to no MR jobs >> being fired. >> It has a dependency on Hive metastore though , through which it derives >> the mechanism to execute the queries directly on source files. >> The only flip side I found was the absence of complex SQL syntax that >> means creating a lot of intermediate tables for little complicated >> calculations (and imho , all calculations become complex sooner than we >> intend them to ) >> >> regards >> Devopam >> >> On Tue, Feb 3, 2015 at 10:30 AM, Samuel Marks <samuelma...@gmail.com> >> wrote: >> >>> Alexander: So would you recommend using Phoenix for all but those kind >>> of queries, and switching to Hive+Tez for the rest? - Is that feasible? >>> >>> Checking their documentation, it looks like it just might be: >>> https://cwiki.apache.org/confluence/display/Hive/HBaseIntegration >>> >>> There is some early work on a Hive + Phoenix integration on GitHub: >>> https://github.com/nmaillard/Phoenix-Hive >>> >>> Saurabh: I am sure there are a variety of very good non open-source >>> products on the market :) - However in this thread I am only looking at >>> open-source options. Additionally I am planning on open-sourcing this >>> project I am building using these tools, so it makes even more sense that >>> the entire toolset and their dependencies are also open-source. >>> >>> Best, >>> >>> Samuel Marks >>> http://linkedin.com/in/samuelmarks >>> >>> On Tue, Feb 3, 2015 at 2:33 PM, Saurabh B <saurabh.wri...@gmail.com> >>> wrote: >>> >>>> This is not open source but we are using Vertica and it works very >>>> nicely for us. There is a 1TB community edition but above that it costs >>>> money. >>>> It has really advanced SQL (analytical functions, etc), works like an >>>> RDBMS, has R/Java/C++ SDK and scales nicely. There is a similar option of >>>> Redshift available but Vertica has more features (pattern matching >>>> functions, etc). >>>> >>>> Again, not open source so I would be interested to know what you end up >>>> going with and what your experience is. >>>> >>>> On Mon, Feb 2, 2015 at 12:08 AM, Samuel Marks <samuelma...@gmail.com> >>>> wrote: >>>> >>>>> Well what I am seeking is a Big Data database that can work with Small >>>>> Data also. I.e.: scaleable from one node to vast clusters; whilst >>>>> maintaining relatively low latency throughout. >>>>> >>>>> Which fit into this category? >>>>> >>>>> Samuel Marks >>>>> http://linkedin.com/in/samuelmarks >>>>> >>>> >>>> >>> >> >> >> -- >> Devopam Mittra >> Life and Relations are not binary >> > -- Devopam Mittra Life and Relations are not binary
