Jumia is a expertise firm born in 2012, current in 14 African nations, with its foremost headquarters in Lagos, Nigeria. Jumia is constructed round a market, a logistics service, and a cost service. The logistics service allows the supply of packages by a community of native companions, and the cost service facilitates the funds of on-line transactions inside Jumia’s ecosystem. Jumia is current in NYSE and has a market cap of $554 million.
On this publish, we share a part of the journey that Jumia took with AWS Skilled Companies to modernize its information platform that ran beneath a Hadoop distribution to AWS serverless based mostly options. A few of the challenges that motivated the modernization had been the excessive value of upkeep, lack of agility to scale computing at particular instances, job queuing, lack of innovation when it got here to buying extra trendy applied sciences, complicated automation of the infrastructure and functions, and the lack to develop domestically.
Resolution overview
The fundamental idea of the modernization mission is to create metadata-driven frameworks, that are reusable, scalable, and ready to answer the completely different phases of the modernization course of. These phases are: information orchestration, information migration, information ingestion, information processing, and information upkeep.
This standardization for every part was thought of as a strategy to streamline the event workflows and reduce the danger of errors that may come up from utilizing disparate strategies. This additionally enabled migration of various sorts of knowledge following the same method whatever the use case. By adopting this method, the info dealing with is constant, extra environment friendly, and extra easy to handle throughout completely different initiatives and groups. As well as, though the use instances have autonomy of their area from a governance perspective, on prime of them is a centralized governance mannequin that defines the entry management within the shared architectural parts. Importantly, this implementation emphasizes information safety by implementing encryption throughout all providers, together with Amazon Easy Storage Service (Amazon S3) and Amazon DynamoDB. Moreover, it adheres to the precept of least privilege, thereby enhancing total system safety and decreasing potential vulnerabilities.
The next diagram describes the frameworks that had been created. On this design, the workloads within the new information platform are divided by use case. Every use case requires the creation of a set of YAML information for every part, from information migration to information circulation orchestration, and they’re mainly the enter of the system. The output is a set of DAGs that run the particular duties.
Within the following sections, we focus on the targets, implementation, and learnings of every part in additional element.
Information orchestration
The target of this part is to construct a metadata-driven framework to orchestrate the info flows alongside the entire modernization course of. The orchestration framework supplies a sturdy and scalable resolution that has the next capacities: dynamically create DAGs, combine natively with non-AWS providers, permit the creation of dependencies based mostly on previous executions, and add an accessible metadata era per every execution. Subsequently, it was determined to make use of Amazon Managed Workflows for Apache Airflow (Amazon MWAA), which, by the Apache Airflow engine, supplies these functionalities whereas abstracting customers from the administration operation.
The next is the outline of the metadata information which might be supplied as a part of the info orchestration part for a given use case that performs the info processing utilizing Spark on Amazon EMR Serverless:
proprietor: # Use case proprietor
dags: # Record of DAGs to be created for this use case
- identify: # Use case identify
kind: # Kind of DAG (might be migration, ingestion, transformation or upkeep)
tags: # Record of TAGs
notification: # Defines notificacions for this DAGs
on_success_callback: true
on_failure_callback: true
spark: # Spark job data
entrypoint: # Spark script
arguments: # Arguments required by the Spark script
spark_submit_parameters: # Spark submit parameters.
The thought behind all of the frameworks is to construct reusable artifacts that allow the event groups to speed up their work whereas offering reliability. On this case, the framework supplies the capabilities to create DAG objects inside Amazon MWAA based mostly on configuration information (YAML information).
This specific framework is constructed on layers that add completely different functionalities to the ultimate DAG:
- DAGs – The DAGs are constructed based mostly on the metadata data supplied to the framework. The information engineers don’t have to jot down Python code as a way to create the DAGs, they’re routinely created and this module is in command of performing this dynamic creation of DAGs.
- Validations – This layer handles YAML file validation as a way to stop corrupted information from affecting the creation of different DAGs.
- Dependencies – This layer handles dependencies amongst completely different DAGs as a way to deal with complicated interconnections.
- Notifications – This layer handles the kind of notifications and alerts which might be a part of the workflows.
One side to think about when utilizing Amazon MWAA is that, being a managed service, it requires some upkeep from the customers, and it’s necessary to have a very good understanding of the variety of DAGs and processes that you just’re anticipated to have as a way to fine-tune the occasion and procure the specified efficiency. A few of the parameters that had been fine-tuned in the course of the engagement had been core.dagbag_import_timeout, core.dag_file_processor_timeout, core.min_serialized_dag_update_interval, core.min_serialized_dag_fetch_interval, scheduler.min_file_process_interval, scheduler.max_dagruns_to_create_per_loop, scheduler.processor_poll_interval, scheduler.dag_dir_list_interval, and celery.worker_autoscale.
One of many layers described within the previous diagram corresponds to validation. This was an necessary element for the creation of dynamic DAGs. As a result of the enter to the framework consists of YML information, it was determined to filter out corrupted information earlier than making an attempt to create the DAG objects. Following this method, Jumia might keep away from undesired interruptions of the entire course of. The module that truly builds DAGs solely receives configuration information that comply with the required specs to efficiently create them. In case of corrupted information, data concerning the precise points is logged into Amazon CloudWatch in order that builders can repair them.
Information migration
The target of this part is to construct a metadata-driven framework for migrating information from HDFS to Amazon S3 with Apache Iceberg storage format, which entails the least operational overhead, supplies scalability capability throughout peak hours, and ensures information integrity and confidentiality.
The next diagram illustrates the structure.
Throughout this part, a metadata-driven framework inbuilt PySpark receives a configuration file as enter in order that some migration duties can run in an Amazon EMR Serverless job. This job makes use of the PySpark framework because the script location. Then the orchestration framework described beforehand is used to create a migration DAG that runs the next duties:
- The primary activity creates the DDLs in Iceberg format within the AWS Glue Information Catalog utilizing the migration framework inside an Amazon EMR Serverless job.
- After the tables are created, the second activity transfers HDFS information to a touchdown bucket in Amazon S3 utilizing AWS DataSync to sync buyer information. This course of brings information from all of the completely different layers of the info lake.
- When this course of is full, a 3rd activity converts information to Iceberg format from the touchdown bucket to the vacation spot bucket (uncooked, course of, or analytics) utilizing once more an alternative choice of the migration framework embedded in an Amazon EMR Serverless job.
Information switch efficiency is healthier when the scale of the information to be transferred is round 128–256 MB, so it’s really useful to compress the information on the supply. By decreasing the variety of information, metadata evaluation and integrity phases are diminished, rushing up the migration part.
Information ingestion
The target of this part is to implement one other framework based mostly on metadata that responds to the 2 information ingestion fashions. A batch mode is chargeable for extracting information from completely different information sources (resembling Oracle or PostgreSQL) and a micro-batch-based mode extracts information from a Kafka cluster that, based mostly on configuration parameters, has the capability to run native streams in streaming.
The next diagram illustrates the structure for the batch and micro-batch and streaming method.
Throughout this part, a metadata-driven framework builds the logic to carry information from Kafka, databases, or exterior providers, that will probably be run utilizing an ingestion DAG deployed in Amazon MWAA.
Spark Structured Streaming was used to ingest information from Kafka subjects. The framework receives configuration information in YAML format that point out which subjects to learn, what extraction processes needs to be carried out, whether or not it needs to be learn in streaming or micro-batch, and wherein vacation spot desk the knowledge needs to be saved, amongst different configurations.
For batch ingestion, a metadata-driven framework written in Pyspark was applied. In the identical manner because the earlier one, the framework obtained a configuration in YAML format with the tables to be migrated and their vacation spot.
One of many elements to think about in this kind of migration is the synchronization of knowledge from the ingestion part and the migration part, in order that there isn’t any lack of information and that information just isn’t reprocessed unnecessarily. To this finish, an answer has been applied that saves the timestamps of the final historic information (per desk) migrated in a DynamoDB desk. Each kinds of frameworks are programmed to make use of this information the primary time they’re run. For micro-batching use instances, which use Spark Structured Streaming, Kafka information is learn by assigning the worth saved in DynamoDB to the startingTimeStamp parameter. For all different executions, precedence will probably be given to the metadata within the checkpoint folder. This manner, you can also make certain ingestion is synchronized with the info migration.
Information processing
The target on this part was to have the ability to deal with updates and deletions of knowledge in an object-oriented file system, so Iceberg is a key resolution that was adopted all through the mission as delta lake information due to its ACID capabilities. Though all phases use Iceberg as delta information, the processing part makes intensive use of Iceberg’s capabilities to do incremental processing of knowledge, creating the processing layer utilizing UPSERT utilizing Iceberg’s skill to run MERGE INTO instructions.
The next diagram illustrates the structure.
The structure is much like the ingestion part, with simply modifications to the info supply to be Amazon S3. This method accelerates the supply part and maintains high quality with a production-ready resolution.
By default, Amazon EMR Serverless has the spark.dynamicAllocation.enabled parameter set to True. This selection scales up or down the variety of executors registered inside the software, based mostly on the workload. This brings loads of benefits when coping with various kinds of workloads, but it surely additionally brings concerns when utilizing Iceberg tables. For example, whereas writing information into an Iceberg desk, the Amazon EMR Serverless software can use a lot of executors as a way to pace up the duty. This may end up in reaching Amazon S3 limits, particularly the variety of requests per second per prefix. Because of this, it’s necessary to use good information partitioning practices.
One other necessary side to think about in these instances is the article storage file structure. By default, Iceberg makes use of the Hive storage structure, however it may be set to make use of ObjectStoreLocationProvider. By setting this property, a deterministic hash is generated for every file, with a hash appended instantly after write.information.path. This will significantly reduce throttle requests based mostly on object prefix, in addition to maximize throughput for Amazon S3 associated I/O operations, as a result of the information written are equally distributed throughout a number of prefixes.
Information upkeep
When working with information lake desk codecs resembling Iceberg, it’s important to have interaction in routine upkeep duties to optimize desk metadata file administration, stopping a lot of pointless information from accumulating and promptly eradicating any unused information. The target of this part was to construct one other framework that may carry out these kinds of duties on the tables inside the information lake.
The next diagram illustrates the structure.
The framework, in addition to the opposite ones, receives a configuration file (YAML information) indicating the tables and the record of upkeep duties with their respective parameters. It was constructed on PySpark in order that it might run as an Amazon EMR Serverless job and might be orchestrated utilizing the orchestration framework similar to the opposite frameworks constructed as a part of this resolution.
The next upkeep duties are supported by the framework:
- Expire snapshots – Snapshots can be utilized for rollback operations in addition to time touring queries. Nonetheless, they will accumulate over time and might result in efficiency degradation. It’s extremely really useful to frequently expire snapshots which might be now not wanted.
- Take away previous metadata information – Metadata information can accumulate over time similar to snapshots. Eradicating them frequently can be really useful, particularly when coping with streaming or micro-batching operations, which was one of many instances of the general resolution.
- Compact information – Because the variety of information information will increase, the variety of metadata saved within the manifest information additionally will increase, and small information information can result in much less environment friendly queries. As a result of this resolution makes use of a streaming and micro-batching software writing into Iceberg tables, the scale of the information tends to be small. Because of this, a way to compact information was crucial to boost the general efficiency.
- Onerous delete information – One of many necessities was to have the ability to carry out laborious deletes within the information older than a sure time period. This means eradicating expiring snapshots and eradicating metadata information.
The upkeep duties had been scheduled with completely different frequencies relying on the use case and the particular activity. Because of this, the schedule data for this duties is outlined in every of the YAML information of the particular use case.
On the time this framework was applied, there was no any automated upkeep resolution on prime of Iceberg tables. At AWS re:Invent 2024, Amazon S3 Tables performance has been launched to automatize the upkeep of Iceberg Tables . This performance automates file compaction, snapshot administration, and unreferenced file elimination.
Conclusion
Constructing a knowledge platform on prime of standarized frameworks that use metadata for various elements of the info dealing with course of, from information migration and ingestion to orchestration, enhances the visibility and management over every of the phases and considerably accelerates implementation and growth processes. Moreover, through the use of providers resembling Amazon EMR Serverless and DynamoDB, you possibly can carry all the advantages of serverless architectures, together with scalability, simplicity, versatile integration, improved reliability, and cost-efficiency.
With this structure, Jumia was capable of cut back their information lake value by 50%. Moreover, with this method, information and DevOps groups had been capable of deploy full infrastructures and information processing capabilities by creating metadata information together with Spark SQL information. This method has diminished turnaround time to manufacturing and diminished failure charges. Moreover, AWS Lake Formation supplied the capabilities to collaborate and govern datasets on varied storage layers on the AWS platform and externally.
Leveraging AWS for our information platform has not solely optimized and diminished our infrastructure prices but additionally standardized our workflows and methods of working throughout information groups and established a extra reliable single supply of fact for our information property. This transformation has boosted our effectivity and agility, enabling quicker insights and enhancing the general worth of our information platform.
– Hélder Russa, Head of Information Engineering at Jumia Group.
Take step one in the direction of streamlining the info migration course of now, with AWS.
In regards to the Authors
Ramón DÃez is a Senior Buyer Supply Architect at Amazon Net Companies. He led the mission with the agency conviction of utilizing expertise in service of the enterprise.
Paula Marenco is a Information Architect at Amazon Net Companies, she enjoys designing analytical options that carry mild into complexity, turning intricate information processes into clear and actionable insights. Her work focuses on making information extra accessible and impactful for decision-making.
 Hélder Russa is the Head of Information Engineering at Jumia Group, contributing to the technique definition, design, and implementation of a number of Jumia information platforms that help the general decision-making course of, in addition to operational options, information science initiatives, and real-time analytics.
Pedro Gonçalves is a Principal Information Engineer at Jumia Group, chargeable for designing and overseeing the info structure, emphasizing on AWS Platform and datalakehouse applied sciences to make sure strong and agile information options and analytics capabilities.
