Big Data

[GB9]

Anyone got any real world examples of big data they have actually worked on?

Strikes me as being marketing tulipe.

[DimPrawn]

It's another vague marketing term like "the cloud" meaning a server somewhere.

Evidently "Big Data" is already dead. It's now "Smart Data" and tools such as predictive analytics that provide value.

[fullyautomatix]

Met a cloud company called Cloudtalent or something like that, 4 ex Microsoft employees started it in 2011, it now has 50 employees and grossed £110 million last year. I always thought Cloud was some marketing tulip but it seems there are companies making money on the back of it.

[bobspud]

Yes you are spot on, but on the bright side clearing up the mess is going to make a few of us very rich.

So far the biggest concern for me comes from the fact that all the places I have seen try to use Big Data didn't want to spend any time cleaning up their DB schemas and understanding what they have today. Instead they wanted to go straight to the part where they bung absolutely everything into one big lake and some magic pixy will then find the data and all its meaning instantaneously. Talk about sold a pup.

I had a wonderful client that had a DB that would have fitted on a 3TB raid partition. They wanted to go to MongoDB so I went though the scaling rules and in the end they would have needed 30TB for mongo... Their answer was ah but its going to be cheap storage and ran off to talk to EMC

I have also heard professionals say don't worry we don't need to maintain any relationships because BD will magically find them for us....

SO I think the number of guys talking about it vs understanding it makes it a very dangerous place to be for a CiO

[DimPrawn]

Yes you are spot on, but on the bright side clearing up the mess is going to make a few of us very rich.

So far the biggest concern for me comes from the fact that all the places I have seen try to use Big Data didn't want to spend any time cleaning up their DB schemas and understanding what they have today. Instead they wanted to go straight to the part where they bung absolutely everything into one big lake and some magic pixy will then find the data and all its meaning instantaneously. Talk about sold a pup.

I had a wonderful client that had a DB that would have fitted on a 3TB raid partition. They wanted to go to MongoDB so I went though the scaling rules and in the end they would have needed 30TB for mongo... Their answer was ah but its going to be cheap storage and ran off to talk to EMC

I have also heard professionals say don't worry we don't need to maintain any relationships because BD will magically find them for us....

SO I think the number of guys talking about it vs understanding it makes it a very dangerous place to be for a CiO

Very true. The marketing hype is "Just bung everything into Hadoop, and out comes insightful gems of wisdom".

Hmmmm.

[MrMarkyMark]

Yes you are spot on, but on the bright side clearing up the mess is going to make a few of us very rich.

So far the biggest concern for me comes from the fact that all the places I have seen try to use Big Data didn't want to spend any time cleaning up their DB schemas and understanding what they have today. Instead they wanted to go straight to the part where they bung absolutely everything into one big lake and some magic pixy will then find the data and all its meaning instantaneously. Talk about sold a pup.

I had a wonderful client that had a DB that would have fitted on a 3TB raid partition. They wanted to go to MongoDB so I went though the scaling rules and in the end they would have needed 30TB for mongo... Their answer was ah but its going to be cheap storage and ran off to talk to EMC

I have also heard professionals say don't worry we don't need to maintain any relationships because BD will magically find them for us....

SO I think the number of guys talking about it vs understanding it makes it a very dangerous place to be for a CiO

+1 saw all this many years ago with Netezza.

[GB9]

Thank Heaven it's not just me then.

[d000hg]

Big Data is surely simply when the volume of data/rate you get it in at exceeds the capability of a traditional DB tool?

Quite real but as most things, gets overused as a term. I don't know how many companies are really working with data on this scale but if you want to treat the internet as a data feed I suppose it can happen quite easily!

[GB9]

Big Data is surely simply when the volume of data/rate you get it in at exceeds the capability of a traditional DB tool?

Quite real but as most things, gets overused as a term. I don't know how many companies are really working with data on this scale but if you want to treat the internet as a data feed I suppose it can happen quite easily!

Depending on what you read volume is just one element.

As the guys have alluded to in earlier posts, it can also include breadthe of data. I've heard several instances where clients have been told they can join anything and everything and the correct result will magically pop out. Usually something plops rather than pops.

And many conventional db will manage more data than most organisations have, although there are a few instances where volume does become an issue

In general I would regard the term Big Data as marketing hype with organisations being sold a dream that they could easily crack with tools other than hadoop etc.

[stek]

Anyone got any real world examples of big data they have actually worked on?

Have you tried MF's tailor?