My main use case for TigerGraph involves finding default patterns in cryptocurrency transactions, where my company received a project that required us to identify those patterns.
A specific example of how I used TigerGraph to find patterns in cryptocurrency transactions was to observe multi-hopping in a limited time frame, where the maximum transaction was happening from one person to multiple people, which is a fraud detection pattern indicating a large amount from one person going to different people in a very small amount of time.
Another use case I had was to observe the amount accumulation with the top five people in a given time frame. As cryptocurrency data is increasing day by day, we took three months of data and inserted it into TigerGraph using CSVs rather than any API to identify the top five people accumulating the most amount, utilizing a heap accumulator in that case.
The best features TigerGraph offers in my experience include its ability to handle large data sets, which made client handling much easier as visualizing that data in table form was very tiring and not at all effective. TigerGraph provided a great experience in data visualization.
I appreciated TigerGraph's visualization because it was easy to see when one person accumulates an amount that goes to one hundred or more people, as the edges and vertices displayed relationships in an amazing way, making it very easy to visualize transactions and patterns. We could change colors, names, and icons according to our requirements, which was very helpful.
Regarding additional features, writing queries was also very interesting, as the queries and accumulators used in TigerGraph made changing the requirements from customers very easy because G-SQL is, in my opinion, quite easy to grasp.
TigerGraph has positively impacted my organization by providing better visualization and transaction patterns compared to our original database, BigQuery, which did not meet client expectations in data representation.
Specific outcomes that show how TigerGraph helped our organization include saving time, as BigQuery was costly in terms of data querying. TigerGraph allowed us to write G-SQL queries and filter out data, definitely reducing manual work and helping us with cost optimization.
TigerGraph can be improved by providing more tutorials about G-SQL. With experience, I have increased my skills for G-SQL, but I found limited tutorials online, so I would really want to see an improvement in G-SQL tutorials or proper documentation.
I have used TigerGraph for about two years.
I did not find any fault regarding TigerGraph's stability.
As for scalability, I did not find any slowdowns or challenges. As the dataset increased, it performed better, and I found no inconsistencies.
I previously used Google Cloud's BigQuery, and we switched because querying data there was very costly, and its table format did not make sense for the relationships of transactions.
I have seen a return on investment with TigerGraph. It is much cheaper than using BigQuery, saving time because we can see patterns right in front of our eyes without needing to go back and change previous queries, thus requiring fewer employees as only a small team worked on TigerGraph.
I do not remember the other software that displays vertices and edges as relationships, but the volume of data we had made TigerGraph the best option due to its architecture and other technical aspects, making it much faster.
For those looking into using TigerGraph, I would advise that understanding the G-SQL part is crucial because without that, choosing TigerGraph would not be beneficial. However, once you grasp G-SQL, which I think is easy, you can master TigerGraph as a developer.
Overall, I think it was good to use TigerGraph, and I had a nice experience. I have given this review a rating of eight out of ten.
