Refactor for Lower Memory

At Rubrik, this class reads a backup snapshot file in fixed-size chunks and computes a checksum for each chunk before it's written to cloud storage. It's used on every backup job, and on large snapshots (hundreds of gigabytes) it has been driving the backup worker's memory usage high enough to trigger OOM kills.

public class ChunkChecksummer {
    public List<byte[]> chunks = new ArrayList<>();
    public List<String> checksums = new ArrayList<>();

    public void process(String filePath, int chunkSize) throws IOException {
        byte[] fileBytes = Files.readAllBytes(Paths.get(filePath));

        int offset = 0;
        while (offset < fileBytes.length) {
            int end = Math.min(offset + chunkSize, fileBytes.length);
            byte[] chunk = Arrays.copyOfRange(fileBytes, offset, end);
            chunks.add(chunk);
            checksums.add(computeChecksum(chunk));
            offset = end;
        }
    }

    private String computeChecksum(byte[] chunk) {
        try {
            MessageDigest digest = MessageDigest.getInstance("SHA-256");
            byte[] hash = digest.digest(chunk);
            StringBuilder sb = new StringBuilder();
            for (byte b : hash) {
                sb.append(String.format("%02x", b));
            }
            return sb.toString();
        } catch (NoSuchAlgorithmException e) {
            throw new RuntimeException(e);
        }
    }
}

Explain what you would change about this class to bring its memory usage down on large snapshot files, and why each change matters.