I/O attribution
Per-statement, per-file, and per-hour read/write breakdown so you know who is touching the disk.
Storage efficiency plan
Cloud storage charges by I/O. We eliminate the reads and writes that never needed to happen, then right-size the volume you actually need.
Diagnosis context
When IOPS climb, the instinct is to buy more storage. The correct move is to ask why the disk is being touched: buffer pool too small, scans that an index would remove, writes from oversized binlog or double-buffering, or backups hammering the same volume as live traffic. We answer that question with data, then cut the cost.
What the fix plan covers
Per-statement, per-file, and per-hour read/write breakdown so you know who is touching the disk.
Buffer pool and cache sizing that turns repeat reads into memory hits instead of disk trips.
Binlog, flush, and doublewrite settings aligned to your durability and cost requirements.
Provisioned IOPS recomputed from the cleaned workload — not from last quarter's worst moment.
Execution path
Baseline storage I/O over a representative window
Attribute reads and writes to their sources
Fix cache sizing and write-path settings
Right-size volumes with measured headroom
Track the savings against latency targets
Measured outcome
A storage bill that reflects your real workload, with latency untouched or improved.
Common Questions
Emergency engagements start with a live diagnostic session — processlist sampling, digest analysis, and parameter review. Stabilization typically lands within one engagement day, with the root cause documented and a monitoring hook to catch recurrence.
Yes, when the strategy matches the change. Instant DDL covers many column operations in MySQL 8; in-place operations fit others; and gh-ost or pt-online-schema-change handle the rest with throttling tied to replica lag. We classify every change before anyone runs it, then rehearse on a production-size replica.
It is an ordering problem, and the fix usually spans both sides. We map the lock graph and specify the transaction re-sequencing your team applies — typically a small, single-commit change. Isolation levels and index design, which are database-side, do most of the heavy lifting.
Because queries filtering or sorting on JSON paths scan without an index. The fix keeps your flexible schema: virtual generated columns over the hot paths, indexes on those columns, and predicate rewrites at the database layer. Your application continues reading the same JSON.
Work with Burj
Tell us what is slow — CPU, queries, migrations, or cost — and we will scope the smallest safe fix.