Using the ESB approach and Data Lake, we reduce the load on systems, eliminate data loss during exchange, simplify changes in IT systems, and speed up the analysis of all data.
You'll also get
Comparison table of
integrations types
.svg)
During point-to-point exchange, data may be lost at thr moment of failure of one of the systems: one system considers the message transmitted when the second one has not received it.
We configure the integration so that when the connection is broken, the message is considered unprocessed and will be transmitted in the next pass.
With the old approach, the source gives the data as it is. The consumer must convert it to his own format. Under this scheme, any change in the source will entail a cascade of changes in all related data streams.
A change in ERP or any other system will not cause improvements to all related systems. If the system changes, it's will only affect source-storage connectors. For all consumers, everything will remain unchanged.
Different systems need the same data. Consumers duplicate requests to the source system, which increases the load. If the company has one system for storing basic data, it will receive a huge number of requests. This is often an old monolithic system that businesses want to replace in the future.
Consumers turn to the storage. The load on the main source is reduced without improving the system itself and increasing resources. When the time comes to replace the old system with several new services, you will need to change the source-storage streams. The changes will not affect many consumers.
To connect multiple consumers of the same type (for example, retailers), an API is often created within the system. At the same time, the burden that consumers create falls on this system. The inaccessibility of the source system will be a problem for all consumers. They won't be able to get data.
We create API connectors not inside the service, but as a separate service. They operate independently of the source system and withstand high loads. You will be able to maintain systems without problems for many consumers.
To transfer an object with the “link” type, you need to make sure that all directories are created in the consumer system. Otherwise, the object will be created broken or the stream will give an error.
Our connectors check whether the system contract is being fulfilled (completeness, correctness).
Only when all conditions are met, the message will be sent to the consumer
Object information can be stored
from several sources. You need to request information on each source and combine it for submission.
Our solution allows you to aggregate object data. For example, collect “orders” and “payments” separately (from different sources) and transfer them as “orders with payments”.
Such a record is not registered for the 1C exchange. The receiver system must know this feature of 1C and calculate changes on its own.
Our connectors can check deletion from the table. If the deletion took place, mark it
in storage. Consumers don't have to do calculations.
1C does not provide a snapshot of current prices. The consumer must determine for himself which price from the entire register (past, present, future) is relevant.
Using storage, we calculate current prices so that any consumer system can pick it up without unnecessary logic.
By implementing ESB, we are creating a common structured enterprise data warehouse: Data Warehouse (DWH).
To such a repository it is easy to connect any analytics and report upload system. All data is ready.
We are building the storage architecture in such a way that generating reports on a large amount of information does not reduce the speed of daily data exchange.
We log the key stages of each stream. If an error occurs that requires a response, you will receive a message on telegram. It will include a description of the bug and a link to details.
You will be able to respond to incidents proactively, rather than to user requests. The support operator will know exactly where and what went wrong. This will help resolve the incident more quickly.
We will deploy a monitoring system from scratch. Or we will set up monitoring in your infrastructure.
With our approach, there is no single point of failure in architecture. A data bus (ESB) is a collection of microservices that are not connected to each other.
ESB components are independent of each other and can run on different servers in different locations.
A failure to exchange either system affects only that system. Everything else is still working.
We are confident in our approach. Tell us about the problem you need to solve. In the documents, we will indicate the result you will receive, the cost and the period (usually 1-2 months).
If you are unhappy with the result after the implementation, you don't have to sign the act and not pay for the work.
.svg)
We use Opensource solutions, so our customers reduce license fees without the risk of restrictions from legislation in different countries.
A simple and flexible low-code platform that is part of Salesforce with revenues of more than $30 million per year. We're using the Community Edition.
An open source, web-based DevOps lifecycle tool. More than 30 million registered users.
A powerful open-source object-relational database system that has been actively developed for over 35 years.
Enterprise security, surveillance, and search solutions built on the Elasticsearch platform used by thousands of companies.
A system for visualizing and analyzing information that allows you to work “out of the box” with a wide range of data sources.
We assessed how three popular types of integrations affect the quality of the IT circuit. The result is presented in a compact comparison table. Submit a request and get the file for free.
The calculator calculates using an accurate but simplified formula. The scope of work for your project and the final cost may vary. The final calculation will be made by your personal manager.
The calculator calculates using an accurate but simplified formula. The scope of work for your project and the final cost may vary. The final calculation will be made by your personal manager.
Personal managers will contact you
Email:
Telegram:
.svg)
