Move Solutions, the French leader in tool-based data migration, will commit to results for your:
- Migration to a new software package or application,
- Data fusion (consolidation of different production instances),
- Data reliability,
- Creation of individual and corporate data repositories with deduplication.
We bring you a tried-and-tested approach and tools, expertise in managing these projects, and many references on large-scale projects carried out for major customers.
What is Data Migration?
When does a data migration project come into play?
Data migration projects are encountered in a various range of situations, such as..:
- Implementation of a software package or new development
- Convergence of Information Systems to rationalize the IS and reduce costs.
Your data migration project involves transporting your data from your old environment to your new system. The new system has a completely different data model and management rules.
Within a limited timeframe and budget, you need to feed the new data structures according to their own integrity rules. The objective is not so much to migrate the Source data as to ensure that the Target System functions correctly with the migrated data. There’s a big difference!
A Data Migration is a Project in itself
Migrating your data to your new system or software package, without disrupting your business, is a project in its own right, and its complexity is often underestimated:
- The source system is often only handled by a small number of “experts”. In most cases, these are already critical resources on the project to implement the target solution,
- The target is moving due to adaptations made to the target data model as part of the process of reducing functional gaps, often carried out using an agile approach.
- By design, new systems are less permissive. They require high data quality, as well as information that may not have been managed, or may have been poorly managed, in the source system. A costly data enrichment and reliability project is therefore necessary, involving a large number of project managers.
- The switchover must take place within a weekend, whatever the volume of data. The performance of the migration chain is therefore critical.
The challenge of iteration
Data migration is often illustrated using referential data such as surname, first name and address. Taking such examples leads us to greatly underestimate the complexity of migration. If everything were so simple, migrating data would simply involve copying it from one database to another. The reality is quite different, and Pareto’s law applies: 80% of the difficulties will come from 20% of the data to be migrated.
Complex migration problems are to be found in calculated data and date sequences.
Migrating a portfolio of bank loans, for example, will require adjustments to the parameters. It will be necessary to ensure that future instalments are calculated in the same way on the Source and Target. All this, despite the fact that the calculation engines are different and may generate discrepancies. Discrepancies that will inevitably give rise to questions or complaints from customers…
To identify discrepancies, you have to try. This leads to an iterative process, which is the only way to converge onto the right migration rule.
If you don’t plan for this from the very start of the project, you’ll run into major difficulties.
So it’s crucial to be able to iterate quickly and well. In other words, you need to be able to implement changes to migration rules quickly and without regression. This is the key to a successful migration project.
Data migration is not a development project like any other, and cannot be carried out efficiently using standard development resources.
Generally speaking, when it comes to Data Migration
Even if software publishers tell you otherwise, ETL or SQL are not suited to large volumes and rapid iteration cycles. Moreover, these technologies don’t produce the technical and functional indicators needed to verify that Lavoisier’s law also applies to data migration: “Nothing is lost, nothing is created, everything is transformed”.
On many occasions, we have been called upon to get out of the rut of projects carried out with well-known ETLs that required extravagant machine resources for poor performance.
Conventional development languages are not productive enough, and don’t allow you to keep control of your specifications. In practice, after a few iterations, the rules are only defined in the code and in the developer’s head, which then becomes a bottleneck and a risk for the project.
For all these reasons, over the years we have developed a dedicated technology based on code generation from specifications.
Our Data Migration offer
Move Solutions’ approach to data migration is supported by dedicated technology. At the heart of the process, a high-performance tool structures the work and optimizes the workload. This enables the teams to keep control of the migration rules, which are always up to date and shared with all those involved in the project.
Code is generated from the rules. It is therefore standardized, reliable and high-performance, and implemented very quickly after the rule has been elaborated or modified.
In this way, we concentrate on the key elements: the business rules.
Our approach ensures that project managers have a business vision of the migration through dashboards and business reports. This drives acceptance tests, data reliability and anomaly analysis.
We take charge of the data migration to your new IS, on a fixed-price basis and with a commitment to results. We also provide to all project players methodological support, consulting and technology. And we do this for all those involved in the project, so as to reduce costs and deadlines.
Steering: a key point in a Data Migration Project
This also has a strong impact on the quality of data migration project management. Our RECODE DATA Workbench provides the raw material for this steering. It enables us to monitor the progress of specifications and outstanding issues for each target data item.
At the start of each phase, we draw up a forecast progress graph. We monitor actual vs. forecast progress. Deviations are detected early and corrective action can be taken without delay. A crucial point is the management of outstanding items. Because they can become blocking factors and have an impact on deadlines. Outstanding issues management is an integral part of our workbench.
Two key indicators are tracked by period and by domain:
- % of specification progress (nb of specified items / nb of total items)
- Number of outstanding questions.
Steering during the implementation and support phases
During the implementation phase of the data migration project, we follow-up planned and actual delivery dates, lot by lot. This monitoring is combined with a quality indicator deduced from the number of anomalies in each batch.
In the support phase – which is a crucial phase of the project – change requests caused by the migration and required to ensure correct operation of the target system loaded with the migrated data must be handled and delivered quickly and reliably. We handle anomalies and changes in the same way. But they are counted separately, to be sure of the quality of our service. If the number of open requests is kept under control and the quality is good, the support team is well sized and provides quality services. The project is converging.
A Data Migration project also implies risk management
Risk management is based on the identification of potential risks and associated risk factors. It involves setting up a prevention plan, and regularly measuring the effectiveness of actions and changes in risk factors.
Move Solutions initializes and monitors the risk follow-up plan for data migration projects as soon as the project is launched. To do so, a risk management plan is set up, and the efficiency of the actions taken is periodically evaluated during project committees.
Joint risk analyses may be defined during project review . Risks identified by Move Solutions and having an impact on the project are analyzed during committees.
Internally, risks are recorded and monitored in the monthly report via a risk follow-up report.
Our tools for successful Data Migration projects
Our workbench, developed and constantly improved by our very active R&D, covers the entire data migration cycle.
All information relating to the migration project is gathered in Recode Datalake, providing all project stakeholders with a 360° view.
Applications and Use Cases of Data Migration
Our workbench is particularly suited to :
- Migration to a software package,
- Migration to a new IS
- Creation of a data loader for a software package or application
- Fusion of databases
- Reliability of data
- Deduplication of Personal and Corporate data repositories
We have strong functional skills and many references, particularly in finance (Banking, Insurance, Pensions, Life), but also in Supply Chain and Human Resources
Over the course of our projects, we have also developed functional skills and expertise in business software packages:
- Banking: Sab 2i, Amplitude, Evolan, Basikon, Kelia, O2S, Codix
- P&C Insurance: Guidewire, Eurassur, Cleva,…
- Personal Insurance: Activ’Infinite, Cleva, Graph talk AIA, PWA Atlantis,
Issue
When legal entities merge, or to rationalize your IT production, you need to carry out information system fusion projects. The aim is to bring together in a single production environment data previously processed in several different environments.
Data from all entities must be standardized and deduplicated to ensure coherent operation of the fused system.
This requires :
– harmonization of parameters
– often complex renumbering of all system keys
– deduplication of individuals and corporate persons.
The amount of information to be analyzed and processed is huge, and can be widely automated thanks to our fusion tools, which ensure the security of your project.
Issue
Data quality is crucial. We ensure data quality control with respect to the technical and functional integrity constraints of your information system.
This operation can be carried out :
– As part of a migration project to a new solution,
– As an independent project, to assess the consistency of your information.
Our offer
We implement a data analysis process based on technical and functional integrity control rules.
Any anomalies detected are corrected by means of specific reliability rules, or communicated to users for manual correction.
Data quality is monitored regularly throughout the project.
KYC is a major challenge for organizations.
The Persons repository is at the heart of customer relationship management and risk management.
At a time when CRM systems are becoming the IS control tower ahead of management applications, the reliability of the repository is taking on new importance.
Issue
If the need seems simple to define: “I want to centralize my customer list, contact details and contracts, and create a single repository, mastering that of management applications”, implementation seems particularly complicated for organizations launching this type of project.
The legacy weighs heavily on the project, making it difficult to even fully identify the repositories to be unified.
In groups formed through takeovers and mergers, it is not uncommon for several applications to cover the same functional area, sometimes even with different management teams.
Segmentation by product also often exists, linked to the coverage of management applications and their ability to rapidly implement a new business product.
Although this is a critical point, the problem is not limited to identifying duplicates and building a deduplicated repository. It will be necessary to set up mechanisms for updating this repository, while respecting the silotage constraints of your applications, and sometimes, due to the functional limitations of some applications, keep duplicates… For operational reasons, the solution often involves a meta-repository, mastering the repositories specific to each application.
Our offer
We set up a process for reconciling physical and legal persons from your different data repositories to create a single data repository.
Duplicate detection is ensured by our phonemization and reconciliation algorithms using fuzzy logic and AI, which exploit all the information you have available to ensure the reliability of the result.
We support you in the implementation of your repository by initializing the deduplicated repository and providing the technical components needed to keep it up to date.
In the context of data migration, we can handle the deduplication of individual and corporate data. To make the best possible use of the results, taking into account the particularities and limitations of the target applications, we have developed business meta-models to simplify migration.
Our tools
Our workberch are fitted with advanced deduplication functions based on phonemization algorithms to identify and merge duplicates. We also standardize addresses according to postal standards.
Our workbench provides you with the means to fully industrialize the migration process:
– Visualization and analysis of source data
– Specification and sharing of migration or merge rules
– Generation of migration programs
– Functional control summaries.
Use Cases
Background The SNCF’s purchasing department managed its operations on an MVS, COBOL, DB2 platform with an outdated client-server Easel front. The
…
The Project was based on the development of a suite of tools dedicated to migrate the SiPo application from a
…
Archiving of data and documents from AG2R Group’s Information System Applications intended to be decommissioned Background For many years, the AG2R La
…
