Quickly run large-scale correlations over typed time-series datasets.
Quickly run large-scale correlations over typed time-series datasets.
Customer Reviews
Industry Analyst / Tech Writer in Information Technology and Services
Advanced user of Google Cloud Interference APIPros
Well-detailed documentation including an API reference guide
prices - $0.002 GB/month for Regional class and $0.007 GB/month for Coldline class
Different storage classes for your desired need - Regional (frequent use), Nearline (infrequent use), and Coldline (long-term use)
Highly-reliable - 99.999999999% of object reliability in a year. Your data is always available even if there is a simultaneous loss of two disks
Data storage is available in many regions - North and South America, Europe, Asia, and Australia
Easy and seamless integration with other Google Cloud Services like Compute Engine, Kubernetes Engine, or App Engine
On the ‘Console' tab you're able to try different SDKs for free. It's convenient for developers
One of the most highly-rated free layers around. Users get $300 in free credit once signed up to any Google Could Platform service during the first year. From the second year onwards, there's 5GB of storage available for use
Above from Quora
Quite costly support fee of about $150 per month for Silver class which is the most basic of services
The cost of downloading data from Google Cloud Storage is quite high
The web interface can be intimidating and confusing
Cons:
General lack of managed services, and limited and outdated versions for what they do offer. Currently only has Postgresql and MySQL which are both outdated and severely locked down with less extensions and options open than anywhere else.
Core GCP products like BigQuery, Spanner, Datastore are nice but very opinionated with limited customizations and observability. Any difference in your workflow from the way they're meant to be used will create problems, without any means to improve performance or find out what's happening.
Poor documentation and broken SDKs. Many products seem to be in beta stage for years, which means no SLA guarantees. Pace of development and new features is also behind AWS and Azure.
Several products are “global” which is great when they work but inevitable problems mean everything is down without recourse. Other clouds force you to think in regions which allows you to build multi-region availability if you need it.
Customer sales and support, described in the nicest terms, is absolutely crap. No company is perfect but GCP is notorious for radio silence and constantly changing contacts so beware that you will likely be on your own here unless you're bringing billions in spend - but at the point you wouldn't be reading Quora anyway.
Above from Quora
Index and load a dataset consisting of multiple data sources stored on Google Cloud Storage. Execute Inference queries over loaded datasets, computing relations across matched groups. Unload or cancel the loading of a dataset.
Quite costly support fee of about $150 per month for Silver class which is the most basic of services
The cost of downloading data from Google Cloud Storage is quite high
The web interface can be intimidating and confusing
Cons:
General lack of managed services, and limited and outdated versions for what they do offer. Currently only has Postgresql and MySQL which are both outdated and severely locked down with less extensions and options open than anywhere else.
Core GCP products like BigQuery, Spanner, Datastore are nice but very opinionated with limited customizations and observability. Any difference in your workflow from the way they're meant to be used will create problems, without any means to improve performance or find out what's happening.
Poor documentation and broken SDKs. Many products seem to be in beta stage for years, which means no SLA guarantees. Pace of development and new features is also behind AWS and Azure.
Several products are “global” which is great when they work but inevitable problems mean everything is down without recourse. Other clouds force you to think in regions which allows you to build multi-region availability if you need it.
Customer sales and support, described in the nicest terms, is absolutely crap. No company is perfect but GCP is notorious for radio silence and constantly changing contacts so beware that you will likely be on your own here unless you're bringing billions in spend - but at the point you wouldn't be reading Quora anyway.
Above from Quora