

During file upload the dataset is processed in memory and not written to disk.
When the upload starts the server generates a cryptographically strong pseudorandom AES-256-CBC Key (24 Bytes)
The client creates “chunks”, sequential byte buffers of the data, each with size 2 MiB
For each chunk we again generate a cryptographically strong pseudorandom initialisation vector (iv)
The raw chunk data is hashed using SHA-256 and then encrypted using the AES key with the initialisation vector
We then create a crc32 checksum of the encrypted chunk.
This encrypted chunk is then written to the file system and the iv, hash and checksum written to the database.
All the asymmetric keys are RSA Keypairs, with at least 4096 bits.
We use RSA with Optimal Asymmetric Encryption Padding (OAEP).
To prevent key exchange attacks all keys are fingerprinted using a SHA-256 hash.
Institute of functional genomics-Statistical Bioinformatics
©2023 · Version 1.12.14
Contact/Impressum · Privacy Policy · Data Policy · Documentation