So im a junior consultant and my current assignment is cleaning

do what you can, read up on what you're not familiar with, succeed or get fired.

if you're fired at least you know it wasnt your fault

>>57638791
There will always be the first time, right, Anon? Maybe your boss just believes you can handle the job.You too should believe in yourself. You can do it!

>>57638791
Go to a book store. Buy like three Dilbert books. Read them all like three times. At that point you should know what to do. Hint: you need to become Wally.

>>57638791
>What would you guys have done in my situation?
watch anime

>>57638819
One more thing. If you are told they have a test server for something, make sure that it's actually a test server, especially since you'll be dealing with financial transactions. My brother managed to (potentially) cost a company a crazy amount of money by fuzzing their "test" server using requests for loans. Despite all signs of the server being a test server (with a test. subdomain even) it wasn't actually a test server. After running that test he got yelled at by their CEO and fired from the contract.

>>57638791
A consultant?
A man who uses your watch to tell you the time?
A man who costs you more than you could make if you ran your business efficiently rather than the way he told you to?
A man who reads the manuals to you over the phone while you try to fix the problems he created?
I just love consultants.

>>57640303
Not all consultants are bad or are total bullshitters, some of them know what they are doing.

>>57638791
>>57638791
You do whatever you have to do to get the job done. It is actually very simple. You already know what the problem is...

The company attempted to build a data warehouse (I don't know what that is exactly, but it sounds like a data storage facility. i.e. it stores records/data [is it digital? as in IT/hard drive storage?]) and it didn't work. Find out what isn't working and fix it, or create a solution to the problem...

Install mongodb or some other meme nosql database

>>57640303
>jealous faggot who isn't skilled enough to do consulting work

>>57641892
>Consulting
>Skill
Googling error messages is not a skill, get a real job and quit being a useless leach.

>>57640638
MongoDB is a web scale database, and doesn't use SQL or JOINs, so it's high-performance.

>>57642191
This is what a consulting is you fucking dimwit.
Doing a "real" job for n years and then starting to share the information what youve gathered.

Please go jump in a well, it will suit you better than the real world.

>>57642248
So you have n years experience googling error messages? You mother must be so proud of you!

>>57642248
That's what a GOOD consultant is. A bad consultant will tell you what you already know and still charge you an arm and a leg. Just like any other facet of business, there are good ones and there are shitty ones that either get weeded out or eventually make enough retarded clones to oversaturate and destroy the ecosystem.

>>57638791
Draw up a concept map of the warehouse architecture and have them implement it. That way if it fails, you can push blame on their implementation did not follow the design.

Welcome to the world of consulting, where you get thrown in at the deep end of the pool on your first day.

Why is the data warehouse a failure? The way you describe it is vague. Is it just crashing, or are there bigger problems with it.
If it is crashing, you can probably figure it out with a decent SQL reference.

More than likely the problems are way bigger than that. Lots of things could have gone wrong: not really understanding what the data warehouse will be used for, selecting the wrong database software, improper technical design, poor implementation, etc.

If you work at a big consulting firm, they probably have some sort of repository of information where you can find information on how to address stuff like this.
If not, you'd better start Googling things like "successful data warehouse" and "failed data warehouse", and use the information you find to come up with a plan.

Whatever you do, run this by your supervisor first before going to your customer.

File: duFLnSuIqWDKzuTE1dPaw5moco5PcJtRwHPkvA0LLwNnA8MBB1DQd-KJCdqeRLjbXaNiuDyMeXZ4T28z-XOa9u6FAbol_IFvsUdW-u4HiIdXPSGHL2cnT2vpf81UbLJJfds-dg=w500-h250-nc.gif (651KB, 500x250px) Image search: [Google]

651KB, 500x250px

>>57642431

>>57638791
The first thing you do is figure out if they tried Kimball or Inmon.

The second thing you do is you get their documentation. If they didn't write any, get a dump of the database schema they used.

If they went with Kimball (90%+ chance) then get the latest edition of the Data Warehouse Lifecycle Toolkit + the other 2 books that come in the set. Read them, look at what they did, figure out what they did wrong.

Unless they tried to be a shitty new age data warehouse company and bought Hadoop's "we're a warehouse too!" bullshit. In that case, you're probably going to have to switch them away from mapreduce and get them on something like Spark, reducing the "warehouse" to a data lake, and building a real data warehouse for them on Vertica or something.