What Really Determines Enterprise Email Migration Performance?
When organisations evaluate email archive migration technology, speed is inevitably one of the first questions asked.
How quickly can we migrate? How much data can we move per day? How long will the project take?
They are important questions, but there is a problem with reducing migration performance to a single technology, software component or framework version.
Enterprise migration performance is the result of an entire architecture working together. Understanding that distinction is essential when comparing migration technologies.
There Is No Single Measure of Migration Speed
A migration platform doesn’t operate in isolation. Data must first be retrieved from a source environment, processed and transformed where necessary, transferred across available infrastructure and ingested into the destination. Every stage can affect throughput.
Among the factors influencing real-world migration performance are:
- Source platform and archive performance
- Health of underlying storage, both in the Source and target
- Retrieval method
- API limitations
- Direct extraction capabilities
- Server specifications
- Network bandwidth and latency
- Parallel processing and batching
- Target ingestion method
- Source and target throttling
- Data size and complexity
- The number and type of items being processed
- Encryption and transformation requirements
This is why migration throughput cannot be attributed to the migration software alone. Where source and target environments are comparable, performance may be similar, but differences across the wider migration architecture can significantly affect the outcome.
Planning a large-scale archive migration?
Talk to us about your source environment, destination, data volumes and performance requirements.
Why Software Framework Versions Don’t Tell the Whole Story
Modern application frameworks can introduce important improvements in areas such as development capabilities, performance and security. But the version number of the underlying framework doesn’t tell you how an enterprise migration application will perform. A newer framework cannot remove a source archive’s retrieval limitations.
It cannot increase the customer’s available network bandwidth. It cannot eliminate Microsoft 365 throttling. And it cannot compensate for inefficient migration architecture or poor application design.
The same principle applies to security. A modern software framework can provide developers with newer security capabilities, but application security ultimately depends on how the application has been engineered, how access and authorisation are managed and the security controls surrounding the complete solution.
Microsoft’s .NET Framework 4.8, for example, remains supported on supported Windows operating systems and continues to receive applicable security updates.
For migration buyers, the more useful question therefore isn’t:
“Which framework version does the product use?”
It is:
“How has the complete migration architecture been designed, secured and proven?”
Retrieval Can Determine the Starting Point
Before anything can be migrated, it needs to be extracted. Different archive platforms provide different ways of accessing data. Some migrations can use direct extraction. Others rely on vendor APIs. Some environments require native platform technologies to interpret data correctly. Each method can have different performance characteristics. A migration platform capable of selecting the appropriate retrieval method for the source environment can therefore have a significant advantage over an architecture constrained to one approach.
Then There Is the Destination
Extracting data quickly is of little value if the destination cannot accept it at the same rate. Microsoft 365 is a good example. Microsoft protects its cloud services through throttling and service limits. These controls can affect how quickly migration workloads are processed regardless of how fast the migration software itself can operate. Effective migration technology therefore needs to manage the destination intelligently rather than simply attempting to send data as quickly as possible.
Throughput without control can create retries, failures and unnecessary pressure on the target environment.
Scale Isn’t Just About Speed
There is another important distinction: Fast isn’t necessarily the same as scalable.
Enterprise scalability is the ability to maintain predictable migration performance as data volumes, users, archives and migration workloads increase.
That requires an architecture capable of efficiently managing:
- Multiple concurrent workloads
- Processing capacity
- Network availability
- Source restrictions
- Target throttling
- Error handing and retries
- Data transformation
A product may perform quickly in a controlled test, but that does not necessarily mean it can maintain the same level of performance when confronted with hundreds of terabytes of complex enterprise data.
True scalability is about maintaining predictable performance as the size and complexity of the migration increase.
Accuracy Has to Remain Part of the Performance Equation
There is little value in moving data faster if doing so compromises its integrity.
Enterprise archive migrations can contain years or decades of corporate information, including records subject to legal discovery, retention requirements and regulatory controls. A successful migration therefore needs to balance throughput with accuracy. Metadata must be preserved, exceptions need to be identified, problematic items need to be understood, migration activity needs to be auditable, and Organisations also need confidence that the data arriving at the destination accurately represents the data held in the source.
How Intelligent Migrator Approaches Performance
Transvault Intelligent Migrator has been developed through 20 years of enterprise migration experience. Rather than assuming every migration behaves in the same way, the architecture supports different retrieval, processing and ingestion methods according to the source, target and project requirements. Performance can be optimised around the complete environment, including infrastructure, network capacity, batching, source limitations, target throttling and the characteristics of the data itself. At the same time, Intelligent Migrator maintains detailed auditing and full Chain of Custody so performance doesn’t come at the expense of visibility or accountability.
Ask Better Questions About Migration Performance
When comparing migration technologies, don’t stop at headline throughput figures or individual technology specifications.
Ask:
- How was the figure measured?
- What source and target were used?
- What volume and type of data was migrated?
- Was the data representative of a real enterprise archive?
- How was target throttling handled?
- Were malformed or problematic items included?
- Was metadata validation performed?
- Could every migrated item be accounted for?
And has the architecture been proven against production enterprise workloads?
Those answers tell you considerably more about migration performance than a software framework version ever will.
Performance Is an Outcome, Not a Specification
Enterprise migration performance is the product of architecture, infrastructure, data quality, source capabilities, destination limitations and intelligent workload management working together. The fastest theoretical migration isn’t necessarily the best migration. The objective is to move enterprise data efficiently while maintaining its integrity, security and auditability from source to destination. That is the standard against which enterprise migration performance should be measured.
Planning a large-scale archive migration?
Talk to us about your source environment, destination, data volumes and performance requirements.