Mailing-List: contact general-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: general@hadoop.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
MIME-Version: 1.0
In-Reply-To: <763307E7-E7FA-43CE-AAE1-9C5A65211AED@nimble.be>
References: <763307E7-E7FA-43CE-AAE1-9C5A65211AED@nimble.be>
From: Ted Dunning <tdunning@maprtech.com>
Date: Tue, 27 Nov 2012 15:17:14 -0800
Message-ID: 
 <CAND0qzuo2vjh6qu+zLac=bK9cA1aDhKOSjeJiX35JtRhwtA=FQ@mail.gmail.com>
Subject: Re: Is Hadoop a good choice for car insurance/telecom provider
 calculators?
To: "general@hadoop.apache.org" <general@hadoop.apache.org>
Content-Type: multipart/alternative; boundary=bcaec54fb7f63fdba104cf824170

--bcaec54fb7f63fdba104cf824170
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

If your computation fits in Excel, then you won't likely see benefits from
Hadoop.

If your computation becomes thousands to millions of times too large for
Excel, then Hadoop might be a great answer.


On Tue, Nov 27, 2012 at 12:39 PM, Gabriel Lozano-Moran <gabriel@nimble.be>w=
rote:

> Hello
>
> I work for a large consumer organisation and we provide as a service for
> our members several calculators where our members can calculate based on
> their profile which is the:
>
> - cheapest energy provider (based on your yearly consumption)
> - cheapest car insurance (based on some criteria such as car brand, model=
,
> age, accident history, =85)
> - cheapest telecommunications provider (based on criteria such as the
> number of text messages sent per month, =85)
>
> For each of the calculators we have data for providers. For example we
> have from the different telecommunications providers all their subscripti=
on
> plans that allows us to calculation for you which would be the cheapest
> provider.
>
> We have right now a custom solution build in .NET (C#) that uses Excel an=
d
> the calculations are performed on several servers.
>
> We are considering on replacing this solution by a new one and we were
> wondering whether switching to Hadoop and Hive ODBC would be a better
> choice for these kind of calculations. Or are there better solutions?
>
> Any help is strongly appreciated.
>
> With kind regards
>
> Gabriel Lozano-Moran
>
>
>

--bcaec54fb7f63fdba104cf824170--