Hi all,

I am evaluating a few Excel-based front end reporting tool for SQL Server Analysis Services. I've evaluated this product from Germany - MIS Plain, and I find it very interesting.

Have anyone had any experience using it? If so, what are your comments?
If not, what Excel-based front end reporting tool would you recommend?

Thanks in advance!

Hi,

there will be a solution from Microsoft, soon... Not as good as Plain, but
it will be free and will be much better than the standard Excel OLAP
browser...


Thomas

<Au>; "Chen Hsiung" <anonymous (AT) discussions (DOT) microsoft.com> schrieb im
Newsbeitrag news:0253D64D-74BD-4B0C-ADD1-00B46D0F5252 (AT) microsoft (DOT) com...
I find it very interesting.

Thomas,

Are you quite confident MS is developing a NEW front end,
Excel based OLAP tool that will work with SQL Server?

Will this OLAP tool be a significant improvement to MS
PTS in OFFICE 11?

Any information you have regarding this new product would
be greatly appreciated.

Thanks for your help,

Brad

Brad,

Microsoft will soon publish Office Add-Ins for Balanced Scorecards, OLAP
(Excel) Reporting and more...

See
http://www.microsoft.com/presspass/p...rprogrampr.asp
for a little more information, the Excel Reporting is currently in Beta...
It looks nice but doesn't do all the tricks Plain is able to do...

HTH


Thomas Pagel, Software4You Planungssysteme GmbH

"Brad" <anonymous (AT) discussions (DOT) microsoft.com> schrieb im Newsbeitrag
news:51d501c3e536$bba6eea0$a101280a (AT) phx (DOT) gbl...

Brad,
Microsoft are releasing one of their "Office Solution Accelerators"
for Excel and OLAP. My understanding is that it will be free for
enterprise licence customers and is part of an extended partner
offering (whatever that means!)

http://www.microsoft.com/office/solu...s/default.mspx

My view is that it will not be as good as the full function Excel
Add-ins, its notoriously difficult to get good fuctionaility,
performance and stability in this area and the guys who have been
doing this for 3 or 4 years and who have many years of development in
this area are ahead. I would say that, as I am one of them at XLCubed,
but you should compare and contrast the respective tools against your
specific requirements when it comes out.

The main players that I see and hear about on a daily basis are

XLCubed, IntelligentApps and MIS

And there are others such as Visual Olap, OlapNavigo, O2Olap etc etc

Prices vary from a couple of hundred dollars to around 750 dollars per
seat.

One thing is for sure with the entrance of a microsoft into the
market, one of the things we come up against all the time goes away -
why do I need anything other than pivot tables.

I expect to see all Full Feature Add-in vendors do a critique of the
Microsoft add-in once it is in the public domain, we certainly will
be.

In terms of free information about excel add-ins

1) This newsgroup has some good old threads (search on any of the
vendor names)
2) http:\\www.johnkeeley.com - John is an Excel Add-in evangelist and
has been for many years who now works for MIS.

Also Nigel Pendse wrote a long paper on them on
http:\\www.olapreport.com which is subscription. I believe you can buy
the paper seperately as well.

Hope that this is helpful

Thanks

Mark Scanlon
XLCubed

"Brad" <anonymous (AT) discussions (DOT) microsoft.com> wrote

I haven't yet seen Microsoft's new Excel offering.
However, I'm inclined to agree with Mark that it is unlikely to be in
the same league as Plain or XLCubed.
It will take Microsoft a few years yet I suspect.

There still remains an important point to remember.
That is, Analysis Services isn't great for planning.
Complex business models are much easier to build with TM1 & Alea OLAP
platforms.
Writeback can still cause problems with AS compared to TM1 & Alea.

So if you want to do serious planning (& most organisations need to)
you may not opt for AS or you may need Alea & AS OLAP platforms.

The advantage Plain has is you can create dynamic reports that
reference both Alea & AS platforms.
To the end user this is seamless.
You get the best of both worlds: AS strength as the OLAP of choice
sitting on top of a datawarehouse; & Alea's strength as a planning
platform.

And AS isn't going to get much better at planning in the Yukon release
either.

Regards,

John

www.johnkeeley.com

John

I am going to have to differ with both yourself and Mark on this and say
that writeback works fine and there are techniques far in advance of both
TM1, Alea and another well know OLAP product as detailed in a previous
thread. For the record, below is a quote from an experienced OLAP user that
REPLACED XLCubed and that comes from one of the environments you mention.

"I have been working with Howard Taylor of O2OLAP closely since installing a
Microsoft Analysis Services based management information solution at my
company. We chose O2OLAP as our data browser because it offered the most
user friendly and flexible interface. It enabled us to create the complex
models we needed with ease and gives the rapid response times expected from
a modern management information system. The views in particular make it
possible to carry out large and complex queries extremely rapidly on cubes
in excess of 10 million records. Anyone looking for an Excel browser for
their Microsoft Analysis Services solution would do well to look at O2OLAP."

There is no reason to be on any other platform other than an Microsoft
Analysis Services platform and why would anyone want to tie their
information and modelling into a platform that is closed with its MDX
equivalent functionality. I do agree with you that things are going to get
better with Yukon and having seen Yukon and asked all the relevant questions
to the right people, all that can be done in the two products you detail can
get done on the Microsoft platform, but with quite a lot of other benefits.

There are also other developments in the market that have been announced
over a week ago that are fairly crucial for solutions and how you approach
them.

Regards
Howard.Taylor@ domain below.
www.o2olap.com


"John Keeley" <duvinrouge (AT) servihoo (DOT) com> wrote

Howard,
It was John who said writeback was not good - not me. Writeback is
fine when you are at leaf level and anybody knows that .. its when you
are at a higher level that you can get into trouble ..

What John agreed with me, was that Microsoft have some catching up to
do in the full function excel add-in market .. a sentiment that I
would have thought that you would have agreed with !!

The architecture of AS means that a lot of data will be written back
to a SQL table if you change a high level aggregated number in a cube
with many dimensions.. now if you have magically found a way to stop
those SQL inserts into the writeback table maybe you should come out
and tell the newsgroup explicitly that rather than just imply that you
have cracked the problem ..

If you are not using AS writeback to do your planning then fine but at
least tell people.

At the end of the day it comes down to architecture and understanding
the architecture in detail, understanding those constraints and then
delivering the appropriate application design against that ..
Writeback is fine in AS when the cube is designed well and the product
that is being used is built to minimise the shortcomings of AS. At the
end of the day - 1 change in a cell in excel can result in many
thousands (if not hundreds of thousands) of inserts into the writeback
table ..

With SP3 the processing of these updates has the potential to be
improved but at the end of the day the inserts are happening .. For
general and very useful details of how to improve writeback
performance with SP3 take a look at ..

http://msdn.microsoft.com/library/de...lwriteback.asp

For someone who knows TM1 which I know you do .. that mythical OLAP
product that people still buy (for quite a lot of money) because its
great at planning! .. the in memory aggregation architecture of TM1 is
better for writeback than AS right now and anyone with even a few
minutes of exposure to the 2 products will tell you that .. The
responsiveness of a high level non leaf update in TM1 is very very
impressive.

We have done a whole lot of work with AS writeback in our new XlCubed
version 2 including using a background thread so that a large workload
can continue without hogging the machine but I would never try and
tell anyone that the AS architecture for writeback is better or even
equivalent than TM1 or a TM1 clone .. because when you understand the
bits and bytes about what is going on it isn't.

It comes down to cost benefit .. if you have a business process which
has planning needs that when you realistically test it does not
perform well with AS then look at other products. If it works OK for
you then thats just great .. Don't just assume its always going to
work as there are many situations where it won't .. which is what you
seem to be implying.

AS is a great product .. but it has its weaknesses and writeback is a
relative weakness .. users need to test planning applications very
carefully with AS and a lot of consultants that I know do not use
writeback within AS and use different methods ..

Answer me this question ..

If you use O2OLAP on the FoodMart Sales cube for 1997 at say the state
of California with all other dimensions being defaulted to the default
member .. what exactly happens ..

a) How long does the writeback take to complete and how many rows get
written to the writeback table
b) Can you cancel the writeback
c) Can you do any other work in the spreadsheet while the other work
is going on.

Assuming that the answer is a LONG LONG time then how would you advise
a customer that wanted to do this to handle this situation.

The other thing that I would think this group might be interested in
is how O2OLAP compares against John Keeleys performance testing ..
which he did about 6 months ago and the details are on
http:\\www.johnkeeley.com.

XLCubed Explorer IntelligentApps 3.5 MIS
Plain 1.2
Query 1 (c.100k cells) 2.2 sec.'s 19.6 sec.'s 8.8
sec.'s
Query 2 (c. 1 million cells) 13.6 sec.'s 228 sec.'s 49 sec.'s

If you wish to repeat these tests yourself here are the details:

Query 1: Foodmart Sales cube - columns: Time.1997 & its descendents,
Measures.Sales Count; rows: Customer.Name

Query 2: Foodmart Sales cube - columns: Customer.City, Gender.Gender,
Measures.Sales Count; rows: Product.Name, Time.Quarter

Given your "Google AD" claim that O2OLAP was the fastest on the market
at the time I am looking forward to seeing your benchmark so that the
new versions of XLCubed and IntelligentApps and MIS Plain can have a
new even faster target to aim for.

I also think that we should encourage people like John who have put a
lot of time into this newsgroup over the years as a user and now as
consultant at MIS.

I think we as vendors owe the newsgroup to provide facts and
information and insights into our architecture and what we have found
to work/not work in our experience of AS not just promote our products
... name specific customer quotes, name people who have done
independent analysis and publish facts .. Our websites are for our
marketing ..

I could list people who have looked at O2OLAP and choosen XLCubed and
bang on about the list of our global customers .. some of whom are
doing global replacements of the XXX product you keep referring to it
... but I don't think this group is particularly interested .. if they
are they can look on websites, read the referenceable quotes, read and
try to understand the collateral and download the products (if they
are allowed!) to have a look to see if they meet their business need.

It really does not take much time to download IA,MIS, XLCubed and
O2OLAP etc etc and look at them and make an informed decision .. each
has its strengths and each may appeal to different people .. There is
no accounting for taste after all !!

Once the microsoft add-in is out and is most likely free then all of
us are going to have to ensure our products represent a significantly
better value proposition and its going to be facts and figures that
will be important in that. Personally I relish the challenge as
challenge is what it will be.

I hope that the newsgroup gets some value out of this post as this is
what the newsgroup is for and when I post that is what my prime
consideration is ..

Thanks

Mark

Mark

higher level that you can get into trouble ..

Correction. There are techniques that can be used for non leaf member input,
but there are also some considerations for this.

thought that you would have agreed with !!

It is going to be interesting what Microsoft will do in the long term. From
O2OLAP's point of view, the creation of a flexible and functional browser
has been part of our business plan, but not never critical.

dimensions.. now if you have magically found a way to stop those SQL inserts
into the writeback table maybe you should come out and tell the newsgroup
explicitly that rather than just imply that you have cracked the problem ..


Our writeback works well and O2OLAP have invested a lot of time and effort
into developing this functionality. We do not intend giving this
functionality away as has effectively been asked by software companies on
more than once occasion now. I do not see other software companies giving
away their investments.

the appropriate application design against that .. Writeback is fine in AS
when the cube is designed well and the product that is being used is built
to minimise the shortcomings of AS. At the end of the day - 1 change in a
cell in excel can result in many thousands (if not hundreds of thousands) of
inserts into the writeback table ..

Agree, but then totally disagree.

at planning! .. the in memory aggregation architecture of TM1 is better for
writeback than AS right now and anyone with even a few minutes of exposure
to the 2 products will tell you that .. The responsiveness of a high level
non leaf update in TM1 is very very impressive.

As you are aware, both architectures have their pros and cons. The TM1
aggregation architecture is quick, but then you can achieve that sort of
speed for AS. There are also different writeback techniques that are
available, each having their merits and being applied for different reasons
in different circumstances. With respects to differences between the two
products there are two areas that TM1 works slightly differently to AS, but
I believe the differences are overcome within Yukon. Re you comment about
updating at non leaf members in TM1, unless things have changed, you could
never do this and effectively had to create a leaf member for the non-leaf
data
member, then rolling this up to the non-leaf data member.

With respects to the rest of the comments on writeback and how people work
with writeback in applications, it is your prerogative to advise as you and
your companies see fit.

about 6 months ago and the details are on

I am aware of this test and will perform this at some stage, but it is not a
priority, nor considered a benchmark. Saying this, if our test proved
faster, then so be it, but there are more important overriding factors that
need to be considered! Speed, together with functionality, is very important
and we rely on our users and resellers for their commentary and continued
development.

I do agree with your comment about encouraging people on this board like
John, who was doing a good job. Being a market commentator is a difficult
job, and most of us know there is always an opportunity to learn something
new. We are all entitled to our opinions, which can be challenged at times.
How we react and deal with differences in opinion can be gleaned from the
response or lack of response. We do not always get it right, and we need
to accept that. Being prepared to fail is part of the learning experience we
all face. Understanding the relationships within the market is also
important
as this helps objectivity for the user as all they are after is a fair
independent
view.

When you buy shares in a company you rely on the financial information of
the company as an opinion as to their fairness has been given by an
independent party, the auditor. Users need to have the same confidence in
commentary for all industries. There is some great commentary within our
market and I will still recommend commentators that I do have a difference
in opinion with as they are an invaluable source of knowledge and ideas.

I do agree with you that users should have the opportunity to see each of
the products on the market.

Howard.Taylor@ domain below
Chartered Accountant (SA)
www.o2olap.com

A correction to the statements below.

From O2OLAP's point of view, the creation of a flexible and functional
browser
has been part of our business plan, but never critical.

Howard

"Howard Taylor [O2OLAP]" <Howard.Taylor (AT) no (DOT) o2olap.com> wrote