ArangoGraph (Managed Service)
ArangoDBReviews from AWS customer
0 AWS reviews
-
5 star0
-
4 star0
-
3 star0
-
2 star0
-
1 star0
External reviews
115 reviews
from
External reviews are not included in the AWS star rating for the product.
Excellent compromise between flexible structure and RDBMS capacities
What do you like best about the product?
The semi-structured and very extensible document model, combined to most of the assets of a hard RDBMS : indexes, joins… plus a few interesting capacities : fulltext indexation, cluster mode. Also, the very active and helpful developers team.
What do you dislike about the product?
Sometimes it feels like some features are still lacking or don't give their full potential yet, but the product is constantly evolving and new features appear every month.
What problems is the product solving and how is that benefiting you?
Our use case is scientific (non-comercial) statistical exploitation of crowdsourced data. Previously using CouchDB, we ran into limitations due to disk space increasing exponentially, and difficulty to perform various complex queries in a reasonable time. ArangoDB allows all that by leveraging most of the powerful concepts of a RDBMS, while keeping a very convenient and extensible semi-sctructured documents orientation. It opens new perspectives for data-mining while keeping a very good read performance thanks to indexes.
Recommendations to others considering the product:
Cluster mode doesn't implement all functionalities yet, and is way slower (at least with community edition). While this is totally understandable, it might be disappointing if one expects it to be "magical scaling". The replication is not as good as CouchDB at the moment.
Good all purpose database with great promise
What do you like best about the product?
Ease of use, nice admin ui where you can test queries (and explain them) and play with arango, many driver options. Overall a good developer experience. It is under active development and there is a github/slack where the arango team responds to issues. AQL is simple and parallels SQL nicely while offering some of its own special things.
What do you dislike about the product?
Some small bugs and quirks. Have upgraded a few times to get bugfixes. I dislike that there are enterprise only features. (I guess you gotta make money somehow.) I also found that under heavy load I experienced seemingly random errors and that my applications could overload the database, locking it up. The java driver didnt have all the features I wanted, namely a connection health check. One pesky frustration is that when you google stuff from the documentation, the top result is often not the current version of the documentation, and then switching versions on the page brings you to the documentation home page.
What problems is the product solving and how is that benefiting you?
Multimodel is nice - arango can be used for different projects with different goals. Also taking advantage of their clustering for redundancy.
Good software, doc and community
What do you like best about the product?
Documentation, tooling, and more generally user experience are the best I had in a long time.
The software has a good default configuration, no bullshit, and it is based on well-tested building blocks.
The community and especially Arango's employee are always very responsible for all kinds of problems.
And of course, pretty decent performances and forgiveful to misuse of some features.
The software has a good default configuration, no bullshit, and it is based on well-tested building blocks.
The community and especially Arango's employee are always very responsible for all kinds of problems.
And of course, pretty decent performances and forgiveful to misuse of some features.
What do you dislike about the product?
Client API are not that good nor efficient. One would have expected a better C++ client than one based on Curl for instance.
Actually, it is so easy to start using and testing Arango that one can miss some important things in the documentation and gives a false impression of bugs/slowness sometime.
Actually, it is so easy to start using and testing Arango that one can miss some important things in the documentation and gives a false impression of bugs/slowness sometime.
What problems is the product solving and how is that benefiting you?
We try to represent a social graph and get out business value out of the links. Traversals, for instance, are something difficult to do efficiently and Arango seems to do a decent job.
Recommendations to others considering the product:
Work on your data model and understand relations.
Amazing product
What do you like best about the product?
The AQL query language is very easy to use, and powerful at the same time
What do you dislike about the product?
Because it's still very new and not yet super popular, there is a severe lack of tools and the community is still very small. But that is very likely to improve over time, as the product itself is truly fantastic.
What problems is the product solving and how is that benefiting you?
Complex Graph queries in an ERP being developed for the Brazilian market
Recommendations to others considering the product:
Assess the complexity and characteristics of your domain before choosing your datastore.
Every domain has a datastore that best matches its needs, and ArangoDB stands on a very strong position in that front, because it's a multi model database, meaning no matter what your domain looks like, ArangoDB is likely ready to support it well.
There is no one size fits all solution though; you have to carefully analyse your domain before making a decision -- people tend to jump into the trendy product, just to find out it was not designed to solve the problem they have.
Every domain has a datastore that best matches its needs, and ArangoDB stands on a very strong position in that front, because it's a multi model database, meaning no matter what your domain looks like, ArangoDB is likely ready to support it well.
There is no one size fits all solution though; you have to carefully analyse your domain before making a decision -- people tend to jump into the trendy product, just to find out it was not designed to solve the problem they have.
ArangoDB: multi-model database
What do you like best about the product?
ArangoDB is a multi-database model that allows you to take full advantage of the flexibility of NoSQL and does so with very good performance. The AQL language is very easy to understand and use. It's really quite friendly with anyone who is already familiar with SQL. ArangoDB also provides good facilities such as the Web UI and CLI tools for database administration.
What do you dislike about the product?
Nothing until now. The database fits well to my current needs.
What problems is the product solving and how is that benefiting you?
Mainly for studies and academic use. A few months ago I used ArangoDB to store data on nutritional information from foods extracted from API's of Brazil and US.
Recommendations to others considering the product:
Try and enjoy it !
Very performant free graph DB
What do you like best about the product?
Very professional graph DB designed from the ground up for multi model support, scaling (e.g., clustering and sharding built in), free open source and enterprise model but even free version has great community and official support.
Very good driver support, Java driver has become really advanced now with async support and direct access to VelicyPack serialization format.
Very good driver support, Java driver has become really advanced now with async support and direct access to VelicyPack serialization format.
What do you dislike about the product?
Proprietary query language AQL, no support for Apache Tinkerbob
Web UI still only has quite basic navigation, visualisation and query assistance (intelligent auto completion)
Web UI still only has quite basic navigation, visualisation and query assistance (intelligent auto completion)
What problems is the product solving and how is that benefiting you?
Using it for our generic sequencing platform, from user and UI level simple configuration storage to storing massive amounts of variant and sample data.
Recommendations to others considering the product:
Didn't check out 3.3 yet, Linux support is great. Watch out for resource usage, now with new RocksDB storage engine much easier on the system resources.
Worth testing
What do you like best about the product?
Graph analysis is very easy, the database helps a lot in connecting different entities rather than bringing them to Spark then try to find those connections, which gives a lot of flexibility and speed boost.
What do you dislike about the product?
No ANSI SQL support, I was hopping for SQL support, that was a disappointing thing to learn a new language. We use CQL at the moment, Spark SQL, and SQL on MySQL, now we have to adapt to AQL which is an unnecessary overhead in my opinion.
What problems is the product solving and how is that benefiting you?
Graph analysis, rather than Spark + Cassandra, yet still experimenting.
ArangoDB the multi-model Database
What do you like best about the product?
Different data models in the same database: Graph DB, Document Store and Key-Value Store with a single language for queries. With Foxx you can create HTTP microservices running inside the database.
What do you dislike about the product?
Nothing for now. Maybe need more community activities in different countries.
What problems is the product solving and how is that benefiting you?
To serve JSON documents and key-value objects in a single database core with a great performance and single language to manage them.
Recommendations to others considering the product:
Try it!!!
An easy way to do complex things
What do you like best about the product?
Document and Graph database at the same time;
It's easy to query the database with AQL;
It's easy to query the database with AQL;
What do you dislike about the product?
Not a big issue, but I think database should keep the connection open like other databases.
What problems is the product solving and how is that benefiting you?
We create an Open Source application called Globomap with an ability to map all our infrastructure and cloud resources, and correlate with each other. So, we can find any resource, no matter the starting point.
Recommendations to others considering the product:
If you coming from another document database, just give it a try. The learning curve is minimal compared to other graph databases.
AQL will help you with complex queries.
AQL will help you with complex queries.
A single database for all my use-cases
What do you like best about the product?
The database allows you to the ability to literally serve any of your use-case and replace any of your current databases. It can act like a schema-less document store or as a graph database, that too without giving up on RDBMS features like Joins.
The query language AQL is especially very powerful.
Besides, the community is great and helpful!
The query language AQL is especially very powerful.
Besides, the community is great and helpful!
What do you dislike about the product?
Not anything specific actually, it is actually great.
The performance of RocksDB storage engine is comparatively lower than the standard mmfiles; But that TBH is not a reasonable thing to expect and would be like comparing apples to oranges because that performance is gained at the cost of resources.
The performance of RocksDB storage engine is comparatively lower than the standard mmfiles; But that TBH is not a reasonable thing to expect and would be like comparing apples to oranges because that performance is gained at the cost of resources.
What problems is the product solving and how is that benefiting you?
We were able to retrieve data quickly, give a search feature and show graphs in our applications - efficiently all using a single database i.e. ArangoDB!
Also, my benchmarking against Neo4j showed results very much in favour of ArangoDB (hence our application served much faster).
Also, my benchmarking against Neo4j showed results very much in favour of ArangoDB (hence our application served much faster).
showing 81 - 90