Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
MIME-Version: 1.0
Message-ID: <pony-cb367aa5aed1991b197e48bcdd89864c2a894a70-3701c4f3f536145b2116aee69f8bcda5819a7ca8@user.cassandra.apache.org>
Subject: Re: Does select queries objects will be allocated in new_heap ??
References: <CAG+7mAsWVtOT50MyBx_2r2Bj36bFQEEMrmauYnR1eoNE6CBPUw@mail.gmail.com> 
From: "Jeff Jirsa"<jjirsa@apache.org>
In-Reply-To: <CAG+7mAsWVtOT50MyBx_2r2Bj36bFQEEMrmauYnR1eoNE6CBPUw@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"
Date: Thu, 06 Jul 2017 05:19:06 -0000
To: <user@cassandra.apache.org>
archived-at: Thu, 06 Jul 2017 05:19:11 -0000


On 2017-07-05 21:50 (-0700), Pranay akula <pranay.akula2323@gmail.com> wrote: 
> Hai,
> 
> I have a small doubt i am not much familiar with java, so when a select or
> insert query is executed against a node does the output of that query will
> be stored in heap ??
> 

SELECT will create a number of short-lived objects on the heap, yes, but it's not like the memtable - there is not a read-side memtable. Most of the objects are transient and should be collected before being promoted to old gen. There are some exceptions - the key and partition caches are counterexamples.


> For insert query the output will stored in Memtables will is part on
> new_heap right ?? does select and other queries also use the new_heap or
> just writes ??
> 

Just writes.


> Is there any document available correlating queries and heap objects.
> 

Very much version dependent, there are some great videos around about the read path - Tyler's is probably as authoritative as you'll find https://www.youtube.com/watch?v=HuDJBTPdaOA


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@cassandra.apache.org
For additional commands, e-mail: user-help@cassandra.apache.org