T
Todd Wright
Guest
Recently, the Progress DataDirect team conducted a webinar to highlight organizations that use APIs and DataDirect to connect to their data sources for improved analytics and standards-based reporting. It’s no secret that APIs have been in use since the early days of computing—whether it was enabling an ATM to look up a customer’s checking account balance or allowing interactive maps to be embedded within websites and devices via Google Maps.
To highlight the enormous impact APIs are having, the DataDirect team recently commissioned a study designed to uncover how APIs are used across industries.
The study determined that:
The survey allowed us to ask our share of questions. But with all that’s going on with APIs, we thought it made sense to allow others to do the same. Enter our Ask Me Anything webinar. The session was a great opportunity for the DataDirect team to spotlight what we’re up to in the space, and how our customers are using DataDirect for data connectivity. With a large audience in attendance, we opened the floor to questions. So many great questions were asked during the webinar, I wanted to highlight the top four we often hear from organizations trialing our software, along with our responses. Here goes…
It can. For example, OpenAI's DALL·E REST API can build a visualization where one can enter a text description of a scene or object and it will generate an image. Another example is the sentiment analysis API from IBM Watson that allows a user to point to a review, or comment on a news story, and respond with sentiment analysis like fear, anger, happiness or sadness. Pretty interesting use cases that will further evolve as artificial intelligence becomes more mainstream.
It is very much dependent on the performance of the REST API itself. Fortunately, we see most APIs are well-designed these days, particularly software-as-a-service commercial APIs. However, DataDirect can do things to maximize the performance. These include the passing of filters so a DataDirect user can capture and leverage any filters the API has, so only the data the client is interested in is requested and passed over the network. Also, if the result set is large enough to come back in separate pages, rather than bringing it back all together in one very large payload, DataDirect will come back in smaller payloads. And finally, DataDirect can set up separate threads. These threads are configurable within DataDirect Autonomous REST Connector, and the REST response can be processed in parallel. Essentially, if you have four threads requesting the larger results set, DataDirect will divide it into fourths and run all four retrievals in parallel. Also, if you have a select statement with an IN clause, DataDirect can divide and conquer with parallel threads. We’ve designed DataDirect to help you maximize performance wherever possible.
When one asks about using an Autonomous REST Connector against Progress OpenEdge, that would mean going against a REST service—usually the Advanced Business Language (ABL)—rather than directly against the OpenEdge database. To go directly against the OpenEdge database, you would use a Progress DataDirect ODBC or JDBC driver and execute SQL against it. In the case of an Autonomous REST Connector, we’re talking about ABL code exposed via a REST service, which DataDirect Autonomous REST Connector would be able to leverage.
This is a common question within cloud environments. The model file being referenced (the REST file) can be retrieved from the file system and via HTTP or HTTPS. It’s as simple as pointing it to the URL.
Hopefully, this post sheds a bit of light on the things you’ve been curious about, but if not, we’re always available to answer your questions. If you want to learn even more, visit the DataDirect Autonomous REST Connector page and catch the full webinar recording.
Continue reading...
To highlight the enormous impact APIs are having, the DataDirect team recently commissioned a study designed to uncover how APIs are used across industries.
The study determined that:
- 78% of organizations used APIs to connect to their data sources
- 60% use internally built APIs for data exchange and data sharing
- 60% use APIs to build business intelligence reports
- 34% use APIs for enabling data mining
The survey allowed us to ask our share of questions. But with all that’s going on with APIs, we thought it made sense to allow others to do the same. Enter our Ask Me Anything webinar. The session was a great opportunity for the DataDirect team to spotlight what we’re up to in the space, and how our customers are using DataDirect for data connectivity. With a large audience in attendance, we opened the floor to questions. So many great questions were asked during the webinar, I wanted to highlight the top four we often hear from organizations trialing our software, along with our responses. Here goes…
1. Can the Autonomous REST Connector from DataDirect work with artificial intelligence REST services?
It can. For example, OpenAI's DALL·E REST API can build a visualization where one can enter a text description of a scene or object and it will generate an image. Another example is the sentiment analysis API from IBM Watson that allows a user to point to a review, or comment on a news story, and respond with sentiment analysis like fear, anger, happiness or sadness. Pretty interesting use cases that will further evolve as artificial intelligence becomes more mainstream.
2. What about the DataDirect Autonomous REST Connector driver's performance with REST APIs?
It is very much dependent on the performance of the REST API itself. Fortunately, we see most APIs are well-designed these days, particularly software-as-a-service commercial APIs. However, DataDirect can do things to maximize the performance. These include the passing of filters so a DataDirect user can capture and leverage any filters the API has, so only the data the client is interested in is requested and passed over the network. Also, if the result set is large enough to come back in separate pages, rather than bringing it back all together in one very large payload, DataDirect will come back in smaller payloads. And finally, DataDirect can set up separate threads. These threads are configurable within DataDirect Autonomous REST Connector, and the REST response can be processed in parallel. Essentially, if you have four threads requesting the larger results set, DataDirect will divide it into fourths and run all four retrievals in parallel. Also, if you have a select statement with an IN clause, DataDirect can divide and conquer with parallel threads. We’ve designed DataDirect to help you maximize performance wherever possible.
3. Does the performance of pagination against a Progress OpenEdge database require a separate page-indexed field—or how can that be set up to be more performant?
When one asks about using an Autonomous REST Connector against Progress OpenEdge, that would mean going against a REST service—usually the Advanced Business Language (ABL)—rather than directly against the OpenEdge database. To go directly against the OpenEdge database, you would use a Progress DataDirect ODBC or JDBC driver and execute SQL against it. In the case of an Autonomous REST Connector, we’re talking about ABL code exposed via a REST service, which DataDirect Autonomous REST Connector would be able to leverage.
4. If my organization is deploying the DataDirect Autonomous REST Connector in a cloud environment using JDBC, but doesn't have a way to write the REST file locally, can I store it elsewhere?
This is a common question within cloud environments. The model file being referenced (the REST file) can be retrieved from the file system and via HTTP or HTTPS. It’s as simple as pointing it to the URL.
Hopefully, this post sheds a bit of light on the things you’ve been curious about, but if not, we’re always available to answer your questions. If you want to learn even more, visit the DataDirect Autonomous REST Connector page and catch the full webinar recording.
Continue reading...