Sybase NNTP forums - End Of Life (EOL)

The NNTP forums from Sybase - forums.sybase.com - are now closed.

All new questions should be directed to the appropriate forum at the SAP Community Network (SCN).

Individual products have links to the respective forums on SCN, or you can go to SCN and search for your product in the search box (upper right corner) to find your specific developer center.

Extent Managed Indexes (Non-B-Tree) Index

10 posts in Product Futures Discussion Last posting was on 2004-04-08 16:59:29.0Z
Ryan Putnam Posted on 2004-01-09 10:04:54.0Z
Sender: 64a7.3ffe6e6b.1804289383@sybase.com
From: Ryan Putnam
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Extent Managed Indexes (Non-B-Tree) Index
X-Mailer: WebNews to Mail Gateway v1.1s
Message-ID: <3ffe7cc6.651b.846930886@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 9 Jan 2004 02:04:54 -0800
X-Trace: forums-1-dub 1073642694 10.22.241.41 (9 Jan 2004 02:04:54 -0800)
X-Original-Trace: 9 Jan 2004 02:04:54 -0800, 10.22.241.41
Lines: 73
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1480
Article PK: 96487

Hi,

I hope that everyone can follow this, as I am not trying to
obfuscate this, but to present a clear mathematically
concise formal definition of what I am proposing.

Your comments are welcome. I have been pondering a new type
of index, I call in an extent managed index. Curious about
its viability.

Assume:
TABLE001 (PCOL1 NUMERIC(20, 0) IDENTITY, COL2 CHAR(6), COL3
INT, COL4 CHAR(3), ...)

COUNT(*) OF TABLE001 = 900,000,000,000

In addition, COL2 by itself qualifies some 20% set of the
table and COL3 qualifies some other partial set of the
table. Now, I would like to query based upon COL2 and COL3
and retrieve the intersection set of the two sets. However,
a B-Tree index is absolutely enormous and unmanageable on a
table of this extreme size.

Hence, some other sort of automaton must arrange the data in
some sort of fashion for retrieval and/or storage.

KNOW:
A distinctive record is located via its page number P an
element of all page numbers of a database, not a segment,
because a table can have its growth started on a new
segment. In addition, its relative offset in bytes on the
page, B an element of the possible byte offsets of a page.

Lets represent all possible page numbers of our table as P0,
P1, ..., Pi and all possible byte offsets in Pi as B0, B1,
.., Bk

Our table can then be represented as:
P0:B0, P0:B1, ..., P0:Bk, P1:Bk, ..., Pi:Bk or simply, we
just call our table Pi:Bk

Now, I do not want to list all elements of COL2 and COL3 so
I will represent all elements of COL2 and COL3 as 2j and 3m
where 2j is an element of COL2 and 3m is and element of
COL3.

Create a extent managed index as such:
EXTENT TABLE MAP
K1 K2 EXT_STRT EXT_FRE_SPACE (0 or 1)
-------------------------------------------------------------
2j 3m x 0 or 1

Extents as pointed to by condition 2j:3m would be extent x
where x is element of 0 or the natural numbers and x = n*8 -
1 or x = 0, where n is an element of the natural numbers.

Extents only contain the RIDS for the condition 2j:3m or
Pi:Bk. So specifying a condition WHERE COL2 = AA AND COL3 =
1 would process as:

LOOP all records matching condition in the EXTENT TABLE
Locate Extent Start Of Qualifying Record
GoTo Extent Start
Traverse RIDS of extent to data pages and get data

Any thoughts, this type of index would be for DSS.

For even more compression, the byte offset Bk can be removed
and just search the entire page for records matching the
condition.

Thanks,
Ryan


Ryan Putnam Posted on 2004-01-15 18:39:03.0Z
Sender: 5b30.4006dd85.1804289383@sybase.com
From: Ryan Putnam
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
X-Mailer: WebNews to Mail Gateway v1.1s
Message-ID: <4006de47.5b40.846930886@sybase.com>
References: <3ffe7cc6.651b.846930886@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 15 Jan 2004 10:39:03 -0800
X-Trace: forums-1-dub 1074191943 10.22.241.41 (15 Jan 2004 10:39:03 -0800)
X-Original-Trace: 15 Jan 2004 10:39:03 -0800, 10.22.241.41
Lines: 8
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1483
Article PK: 96484

The more I think about it, the more I like this also, remove
the byte offset and only store the page that a match exists
or multiple matches exists just once. ASE could than scan
the entire page for records and not have to store multiple
duplicate entries.

Thanks,
Ryan


Arun Posted on 2004-02-26 19:19:16.0Z
From: "Arun" <arun.patnaik@sybase.com>
Newsgroups: sybase.public.ase.product_futures_discussion
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com>
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
Lines: 24
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.50.4922.1500
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4925.2800
NNTP-Posting-Host: arupatw2k.sybase.com
X-Original-NNTP-Posting-Host: arupatw2k.sybase.com
Message-ID: <403e46b4$1@forums-1-dub>
Date: 26 Feb 2004 11:19:16 -0800
X-Trace: forums-1-dub 1077823156 10.22.90.137 (26 Feb 2004 11:19:16 -0800)
X-Original-Trace: 26 Feb 2004 11:19:16 -0800, arupatw2k.sybase.com
X-Authenticated-User: ase1251
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1514
Article PK: 96517

How do you plan to search for the Matching key columns ? Sequential search ?

If yes, then it is going to be pretty inefficient to do the search.
If not you will end up creating more or less what we already have as the
non-clustered index
structure.

In case I missed something, may be if you draw up the representation for an
example data set and that would clarify this further.

--
Thanks
- Arun

<Ryan Putnam> wrote in message news:4006de47.5b40.846930886@sybase.com...
> The more I think about it, the more I like this also, remove
> the byte offset and only store the page that a match exists
> or multiple matches exists just once. ASE could than scan
> the entire page for records and not have to store multiple
> duplicate entries.
>
> Thanks,
> Ryan


Ryan Putnam Posted on 2004-03-02 21:55:13.0Z
Sender: 2a06.404501e5.1804289383@sybase.com
From: Ryan Putnam
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
X-Mailer: WebNews to Mail Gateway v1.1s
Message-ID: <404502e7.2a1b.846930886@sybase.com>
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com><403e46b4$1@forums-1-dub>
X-Original-NNTP-Posting-Host: 10.22.241.42
X-Original-Trace: 2 Mar 2004 13:55:51 -0800, 10.22.241.42
Lines: 35
X-Original-NNTP-Posting-Host: forums-2-dub.sybase.com
X-Original-Trace: 2 Mar 2004 13:53:58 -0800, forums-2-dub.sybase.com
NNTP-Posting-Host: forums-master.sybase.com
X-Original-NNTP-Posting-Host: forums-master.sybase.com
Date: 2 Mar 2004 13:55:13 -0800
X-Trace: forums-1-dub 1078264513 10.22.108.75 (2 Mar 2004 13:55:13 -0800)
X-Original-Trace: 2 Mar 2004 13:55:13 -0800, forums-master.sybase.com
X-Authenticated-User: ngsysop
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1520
Article PK: 96516

Hi,

You bet, let me work on this and I will post. Thanks for
your reply Arun!!!

See ya,
Ryan

> How do you plan to search for the Matching key columns ?
> Sequential search ?
>
> If yes, then it is going to be pretty inefficient to do
> the search. If not you will end up creating more or less
> what we already have as the non-clustered index
> structure.
>
> In case I missed something, may be if you draw up the
> representation for an example data set and that would
> clarify this further.
>
> --
> Thanks
> - Arun
> <Ryan Putnam> wrote in message
> > news:4006de47.5b40.846930886@sybase.com... The more I
> > think about it, the more I like this also, remove the
> > byte offset and only store the page that a match exists
> > or multiple matches exists just once. ASE could than
> scan the entire page for records and not have to store
> > multiple duplicate entries.
> >
> > Thanks,
> > Ryan
>
>


Ryan Putnam Posted on 2004-03-10 19:03:58.0Z
Sender: 3341.404f64f0.1804289383@sybase.com
From: Ryan Putnam
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
X-Mailer: WebNews to Mail Gateway v1.1s
Message-ID: <404f6750.336c.846930886@sybase.com>
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com><403e46b4$1@forums-1-dub>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="-=_forums-2-dub404f6750"
X-Original-NNTP-Posting-Host: 10.22.241.42
X-Original-Trace: 10 Mar 2004 11:06:56 -0800, 10.22.241.42
Lines: 2111
X-Original-NNTP-Posting-Host: forums-2-dub.sybase.com
X-Original-Trace: 10 Mar 2004 11:03:49 -0800, forums-2-dub.sybase.com
NNTP-Posting-Host: forums-master.sybase.com
X-Original-NNTP-Posting-Host: forums-master.sybase.com
Date: 10 Mar 2004 11:03:58 -0800
X-Trace: forums-1-dub 1078945438 10.22.108.75 (10 Mar 2004 11:03:58 -0800)
X-Original-Trace: 10 Mar 2004 11:03:58 -0800, forums-master.sybase.com
X-Authenticated-User: ngsysop
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1527
Article PK: 96531

I am not much of artist, but I have attached visio with some
comments.

Basically, a B-tree will be used to keep track of only the
distinct values for a specified index. This B-tree index
will point to control extent for the unique value. The
control extent would contain the min page # of value in
extent, max page # of value in extent, and storage location
for an extent. Pages in extents are linked via -><- page
ptrs. Much like allpages tables. Page numbers for values
are to be stored only if a page number is not already being
stored for that value. Page numbers with IX storage extents
are to be sorted via the page number of the value being
indexed.

IX on (COLA)
COLA: VALUEA: pgs: 123, 123, 123, 124, 129, 130, etc...

SELECT *
FROM TABLE
WHERE COLA = "VALUEA"
--

From B-tree obtain, location of control extent "if it
exists" for "VALUEA". Lets Assume it is 8A. Now fire off
many processes in parallel to process the IX storage extents
specified by the control extent for VALUEA.

control extent 8A -> 8B
-> 8C
-> 8D
Where A>B>C>D.

Process the data and return it.

Sort Merges via page numbers:

SELECT *
FROM TABLE
WHERE COLA = "VALUEA"
AND COLB = "VALUEB"
AND COLC = "VALUEC"

Get page numbers for all values into sorted lists:
VALUEA | VALUEB | VALUEC
123 123 124
124 124 125
126 126 126

Process the data further for all matching page numbers from
the sort merge join.

Give me a call if you would like to talk further.
Thanks,
Ryan

> How do you plan to search for the Matching key columns ?
> Sequential search ?
>
> If yes, then it is going to be pretty inefficient to do
> the search. If not you will end up creating more or less
> what we already have as the non-clustered index
> structure.
>
> In case I missed something, may be if you draw up the
> representation for an example data set and that would
> clarify this further.
>
> --
> Thanks
> - Arun
> <Ryan Putnam> wrote in message
> > news:4006de47.5b40.846930886@sybase.com... The more I
> > think about it, the more I like this also, remove the
> > byte offset and only store the page that a match exists
> > or multiple matches exists just once. ASE could than
> scan the entire page for records and not have to store
> > multiple duplicate entries.
> >
> > Thanks,
> > Ryan
>
>


Ryan Putnam Posted on 2004-04-01 11:42:42.0Z
Sender: 16ed.406bff41.1804289383@sybase.com
From: Ryan Putnam
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
X-Mailer: WebNews to Mail Gateway v1.1s
Message-ID: <406c002e.16f1.846930886@sybase.com>
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com><403e46b4$1@forums-1-dub><404f6750.336c.846930886@sybase.com>
X-Original-NNTP-Posting-Host: 10.22.241.42
X-Original-Trace: 1 Apr 2004 03:42:38 -0800, 10.22.241.42
Lines: 28
X-Original-NNTP-Posting-Host: forums-2-dub.sybase.com
X-Original-Trace: 1 Apr 2004 03:42:39 -0800, forums-2-dub.sybase.com
NNTP-Posting-Host: forums-master.sybase.com
X-Original-NNTP-Posting-Host: forums-master.sybase.com
Date: 1 Apr 2004 03:42:42 -0800
X-Trace: forums-1-dub 1080819762 10.22.108.75 (1 Apr 2004 03:42:42 -0800)
X-Original-Trace: 1 Apr 2004 03:42:42 -0800, forums-master.sybase.com
X-Authenticated-User: ngsysop
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1532
Article PK: 96533

This could also introduce a feature I'll call planeing.
That is, lets say we create a plane index as such

create plane index (cola, colb, colc).

This would create 3 planes
plane.plane1 for cola
plane.plane2 for colb
plane.plane3 for colc

This would be the same as creating 3 indexes of this type
create plane index(cola)
create plane index(colb)
create plane index(colc)

This would allow for index qualification searching of cola,
colb, and colc.

However, if colb is only used in a conditional statment with
colc, than we could create our original index as such

create plane index(cola, (colb, colc)).
This would create a plane for cola
and a single plane for colb and colc.

Those are my thoughts,
Thanks,
Ryan


Jeff Tallman Posted on 2004-04-04 03:28:26.0Z
From: Jeff Tallman <tallman@sybase.com>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax; CDonDemand-Dom)
X-Accept-Language: en-us, en
MIME-Version: 1.0
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com><403e46b4$1@forums-1-dub><404f6750.336c.846930886@sybase.com> <406c002e.16f1.846930886@sybase.com>
In-Reply-To: <406c002e.16f1.846930886@sybase.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Original-NNTP-Posting-Host: pcp03383846pcs.potshe01.pa.comcast.net
Message-ID: <406f80d6$1@forums-2-dub>
X-Original-Trace: 3 Apr 2004 19:28:22 -0800, pcp03383846pcs.potshe01.pa.comcast.net
Lines: 43
X-Original-NNTP-Posting-Host: forums-2-dub.sybase.com
X-Original-Trace: 3 Apr 2004 19:28:25 -0800, forums-2-dub.sybase.com
NNTP-Posting-Host: forums-master.sybase.com
X-Original-NNTP-Posting-Host: forums-master.sybase.com
Date: 3 Apr 2004 19:28:26 -0800
X-Trace: forums-1-dub 1081049306 10.22.108.75 (3 Apr 2004 19:28:26 -0800)
X-Original-Trace: 3 Apr 2004 19:28:26 -0800, forums-master.sybase.com
X-Authenticated-User: ngsysop
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1533
Article PK: 96535

Check out Sybase IQ's LF, HNG, HG, etc. bit-wise indexing (note - this
is not the simplistic bitmapped indices the Other guys use).

The result is "planing" but on a more usable level - for instance, you
typically end up with an index on each column. The query processor scans
the indexes in parallel and does bitwise comparisons with other results
to project the final rows. Speed is fantastic for DSS queries (10-100x
faster than RAC, DB2 EEE, OLAP, whatever)

What it sounds like you want is Sybase IQ indexes in ASE (now there's a
thought).

Ryan wrote:
> This could also introduce a feature I'll call planeing.
> That is, lets say we create a plane index as such
>
> create plane index (cola, colb, colc).
>
> This would create 3 planes
> plane.plane1 for cola
> plane.plane2 for colb
> plane.plane3 for colc
>
> This would be the same as creating 3 indexes of this type
> create plane index(cola)
> create plane index(colb)
> create plane index(colc)
>
> This would allow for index qualification searching of cola,
> colb, and colc.
>
> However, if colb is only used in a conditional statment with
> colc, than we could create our original index as such
>
> create plane index(cola, (colb, colc)).
> This would create a plane for cola
> and a single plane for colb and colc.
>
> Those are my thoughts,
> Thanks,
> Ryan


Ryan Putnam Posted on 2004-04-04 17:41:01.0Z
Sender: 509f.40703a00.1804289383@sybase.com
From: Ryan Putnam
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
X-Mailer: WebNews to Mail Gateway v1.1s
Message-ID: <40703a9b.50a2.846930886@sybase.com>
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com><403e46b4$1@forums-1-dub><404f6750.336c.846930886@sybase.com> <406c002e.16f1.846930886@sybase.com><406f80d6$1@forums-2-dub>
X-Original-NNTP-Posting-Host: 10.22.241.42
X-Original-Trace: 4 Apr 2004 09:40:59 -0800, 10.22.241.42
Lines: 54
X-Original-NNTP-Posting-Host: forums-2-dub.sybase.com
X-Original-Trace: 4 Apr 2004 09:41:00 -0800, forums-2-dub.sybase.com
NNTP-Posting-Host: forums-master.sybase.com
X-Original-NNTP-Posting-Host: forums-master.sybase.com
Date: 4 Apr 2004 09:41:01 -0800
X-Trace: forums-1-dub 1081096861 10.22.108.75 (4 Apr 2004 09:41:01 -0800)
X-Original-Trace: 4 Apr 2004 09:41:01 -0800, forums-master.sybase.com
X-Authenticated-User: ngsysop
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1534
Article PK: 96536

I will check out your suggestions. Also, I love your idea.
Seems like if Sybase chose to implement the IQ indexes into
ASE that they would have much code they could copy and
splice to potentially save time.

Awesome.

Thanks,
Ryan

>
> Check out Sybase IQ's LF, HNG, HG, etc. bit-wise indexing
> (note - this is not the simplistic bitmapped indices the
> Other guys use).
>
> The result is "planing" but on a more usable level - for
> instance, you typically end up with an index on each
> column. The query processor scans the indexes in parallel
> and does bitwise comparisons with other results to
> project the final rows. Speed is fantastic for DSS queries
> (10-100x faster than RAC, DB2 EEE, OLAP, whatever)
>
> What it sounds like you want is Sybase IQ indexes in ASE
> (now there's a thought).
>
> Ryan wrote:
> > This could also introduce a feature I'll call planeing.
> > That is, lets say we create a plane index as such
> >
> > create plane index (cola, colb, colc).
> >
> > This would create 3 planes
> > plane.plane1 for cola
> > plane.plane2 for colb
> > plane.plane3 for colc
> >
> > This would be the same as creating 3 indexes of this
> > type create plane index(cola)
> > create plane index(colb)
> > create plane index(colc)
> >
> > This would allow for index qualification searching of
> > cola, colb, and colc.
> >
> > However, if colb is only used in a conditional statment
> > with colc, than we could create our original index as
> > such
> > create plane index(cola, (colb, colc)).
> > This would create a plane for cola
> > and a single plane for colb and colc.
> >
> > Those are my thoughts,
> > Thanks,
> > Ryan
>


Carl Kayser Posted on 2004-04-04 18:12:01.0Z
From: "Carl Kayser" <kayser_c@bls.gov>
Newsgroups: sybase.public.ase.product_futures_discussion
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com><403e46b4$1@forums-1-dub><404f6750.336c.846930886@sybase.com> <406c002e.16f1.846930886@sybase.com><406f80d6$1@forums-2-dub> <40703a9b.50a2.846930886@sybase.com>
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
Lines: 27
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2800.1158
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
X-Original-NNTP-Posting-Host: 146.142.33.192
Message-ID: <407041df$1@forums-2-dub>
X-Original-Trace: 4 Apr 2004 10:11:59 -0800, 146.142.33.192
X-Original-NNTP-Posting-Host: forums-2-dub.sybase.com
X-Original-Trace: 4 Apr 2004 10:12:00 -0800, forums-2-dub.sybase.com
NNTP-Posting-Host: forums-master.sybase.com
X-Original-NNTP-Posting-Host: forums-master.sybase.com
Date: 4 Apr 2004 10:12:01 -0800
X-Trace: forums-1-dub 1081098721 10.22.108.75 (4 Apr 2004 10:12:01 -0800)
X-Original-Trace: 4 Apr 2004 10:12:01 -0800, forums-master.sybase.com
X-Authenticated-User: ngsysop
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1535
Article PK: 96538

FYI. This is also a recent ISUG enhancement request: e01_280 - Encapsulate
a mini ASIQ server into ASE . Voting:

Needed: 4
Recommended: 3
Undecided: 4
Not recommended: 0
Not needed: 0

However there have been other (old) requests with both many more votes and a
higher score that Sybase does not plan to implement. (Examples: e01_006 -
Shrink a database and e01_111 - Enable JTC separately from Merge Joins.)

<Ryan Putnam> wrote in message news:40703a9b.50a2.846930886@sybase.com...
> I will check out your suggestions. Also, I love your idea.
> Seems like if Sybase chose to implement the IQ indexes into
> ASE that they would have much code they could copy and
> splice to potentially save time.
>
> Awesome.
>
> Thanks,
> Ryan
> >


Ryan Putnam Posted on 2004-04-08 16:59:29.0Z
Sender: 406e.407576c2.1804289383@sybase.com
From: Ryan Putnam
Newsgroups: sybase.public.ase.product_futures_discussion
Subject: Re: Extent Managed Indexes (Non-B-Tree) Index
X-Mailer: WebNews to Mail Gateway v1.1s
Message-ID: <407576e1.4074.846930886@sybase.com>
References: <3ffe7cc6.651b.846930886@sybase.com> <4006de47.5b40.846930886@sybase.com><403e46b4$1@forums-1-dub><404f6750.336c.846930886@sybase.com> <406c002e.16f1.846930886@sybase.com><406f80d6$1@forums-2-dub> <40703a9b.50a2.846930886@sybase.com><407041df$1@forums-2-dub>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 8 Apr 2004 08:59:29 -0800
X-Trace: forums-1-dub 1081439969 10.22.241.41 (8 Apr 2004 08:59:29 -0800)
X-Original-Trace: 8 Apr 2004 08:59:29 -0800, 10.22.241.41
Lines: 46
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.product_futures_discussion:1536
Article PK: 96537

Thanks for your feedback. I love the fact that Sybase
allows the customers to vote and make feature requests to
drive their products. To me, it indicates that they really
care about all of their customers and their success.

My opinion from working with IBM and Microsoft products and
support is that to those two you are a just a dollar.

I would also like to mention, that I love the stability,
reliability, performance, and decreased time to deployment
that Sybase has provided us. Keep up the good work. I am
really looking forward to some of the products that you and
XcelleNet provide.

Thanks,
Ryan

> FYI. This is also a recent ISUG enhancement request:
> e01_280 - Encapsulate a mini ASIQ server into ASE .
> Voting:
>
> Needed: 4
> Recommended: 3
> Undecided: 4
> Not recommended: 0
> Not needed: 0
>
> However there have been other (old) requests with both
> many more votes and a higher score that Sybase does not
> plan to implement. (Examples: e01_006 - Shrink a database
> and e01_111 - Enable JTC separately from Merge Joins.)
>
>
> <Ryan Putnam> wrote in message
> > news:40703a9b.50a2.846930886@sybase.com... I will check
> > out your suggestions. Also, I love your idea. Seems
> > like if Sybase chose to implement the IQ indexes into
> > ASE that they would have much code they could copy and
> splice to potentially save time. >
> > Awesome.
> >
> > Thanks,
> > Ryan
> > >
>
>