We use Google Cloud BigTable to help us do the following:
1. Store aggregates
2. Build and keep lookup tables
3. Have a storage that can be looked up really fast to find duplicates
4. Given the large no. of txns it can handle, we use it for enriching our data from pipelines thus having rich information in the flows below.
Pros
High throughput
Low latency
Uptime
Fast Key-value lookups
Cons
Global replication
Complete SQL support
Size limit of 1MB
Likelihood to Recommend
It works great for any use case that requires high throughput for reads and writes, for key-value kinda data. This enables you to use it for usecases where you need to retain data for a long term or even short term but then use it in pipelines to enrich data, sync states between backend and frontend or to quickly lookup say a user.
VU
Verified User
C-Level Executive in Engineering (201-500 employees)
Analytics for big data is our biggest use case for Google Cloud's BigTable platform. It is efficient enough to read, handle and analyse TBs of data under unbelievable times.
Majorly, we have a lot of data captured from customer interactions with our applications which is required to be processed and analysed for gaining benefits from those data sets. Now, this data is so huge as it is accumulated every minute from each applications. Our traditional systems cannot process them all that once, and even if we can do slotting and then anaylse them, it will take longer durations and with that new data is ready to be consumed. So as to solve this problem, we started working on BigTable which has huge capabilities and Google's analytical skills. Its efficient and most importantly - fast!
Pros
Analytics: is at Google's heart. No on can beat Google in this space and BigTable is one of its implementation of this. The insights you gain from BigTable are simply usable in your day to day activities and can help you make real difference.
Speed: Processing TBs and PBs of data under minutes needs real efficient platform which is capable of doing much more than just processing data. All this data cannot be processed by a single machine, but rather huge pairs of machines working in conjuction with each other. BigTable's implementation is one of the finest and allows you achieve great speeds!
Interface: is great. Google has segregated required task under logically placed buttons which takes no time by users to understand and get habituated.
Cons
User interface's responsiveness: I understand so much is going on under the hood, but laggyness is acceptable if a workload is running or being processed. In case their is not workload being process, GUI should work blazing fast. I have faced this at times, and this becomes frustrating as well.
Nothing other than this - BigTable is quite efficient platform and does exactly what it is built for.
Likelihood to Recommend
Google Bigtable is ONLY suited for massive data sets which scale PetaBytes and TerraBytes. Anything under this can easily be done via dedicated VMs and open source tools.
Google Bigtable is expensive and shall be used wisely. It should be utilised only where it is well suited else you would simply be wasting dollars and not utilizing its full benefits.
VU
Verified User
Advisor in Information Technology (10,001+ employees)
