Ni nini transaction isolation levels?

Question

Accepted Answer

**Isolation levels** hudhibiti ni kiasi gani transactions za wakati mmoja zinaweza kuona mabadiliko ya kila moja yasiyokamilishwa (au yanayoendelea) — zikibadilishana **mfanano** dhidi ya **utendaji/concurrency**. Isolation ya juu huzuia kasoro zaidi lakini huongeza locking/contention.

## Kasoro za concurrency (kile isolation huzuia)

```text
DIRTY READ        → reading another transaction's UNCOMMITTED changes (which may roll back)
NON-REPEATABLE READ → re-reading a row gives different values (another txn UPDATED it)
PHANTOM READ      → re-running a query returns different ROWS (another txn INSERTED/DELETED)
```

Haya ni matatizo yanayoweza kutokea wakati transactions zinaendesha kwa wakati mmoja — isolation levels hudhibiti ni yapi yanazuiwa.

## Isolation levels nne za kawaida

Kila level ya juu huzuia kasoro zaidi. **Read Committed** ni chaguo-msingi la kawaida (huzuia dirty reads); **Serializable** ni kali zaidi (transactions hutenda kana kwamba zinaendeshwa moja kwa wakati) lakini inayobana zaidi kwa concurrency.

## Kuweka isolation level

```sql
SET TRANSACTION ISOLATION LEVEL SERIALIZABLE;
BEGIN;
  -- ... operations run at the chosen isolation ...
COMMIT;
```

## Uwiano wa msingi

```text
Higher isolation → MORE correctness (fewer anomalies), but MORE locking/contention
                   → lower concurrency, potential slowdowns, more deadlocks
Lower isolation  → MORE concurrency/performance, but risk of anomalies
→ Choose based on the data's needs: financial/critical data may need Serializable;
  most apps use Read Committed (the default) as a balance.
→ Note: implementations vary (MVCC vs locking); the SQL standard defines the levels
  but databases differ in details (Postgres, MySQL/InnoDB, etc.).
```

## Kwa nini ni muhimu

Transaction isolation levels ni maarifa muhimu ya hali ya juu kwa kuelewa **concurrency katika hifadhidata** — wakati transactions nyingi zinaendesha kwa pamoja (kama zinavyofanya katika programu yoyote halisi ya watumiaji wengi), isolation levels huamua uwiano kati ya mfanano wa data na utendaji, hivyo kuzielewa ni thamani kwa kujenga programu za concurrent zilizo sahihi, zenye utendaji.

Dhana za msingi ni **kasoro za concurrency** (dirty reads, non-repeatable reads, phantom reads — matatizo yanayotokea wakati transactions zinaona mabadiliko yanayoendelea ya kila moja) na jinsi **isolation levels nne** (Read Uncommitted, Read Committed, Repeatable Read, Serializable) zinavyozuia zaidi yake hatua kwa hatua.

Kuelewa hili ni muhimu kwa usahihi: kuchagua isolation level ya chini sana kunaweza kuruhusu kasoro zinazoharibu data au kutoa matokeo yasiyo sahihi (mfano kusoma data isiyokamilishwa inayobatilishwa baadaye, au usomaji usiofanana), wakati kuchagua ya juu sana kunapunguza concurrency na utendaji.

**Uwiano wa msingi** — isolation ya juu maana yake usahihi zaidi lakini locking/contention zaidi (concurrency ya chini, kupungua kwa kasi kunakowezekana na deadlocks), isolation ya chini maana yake utendaji zaidi lakini hatari ya kasoro — ni ufahamu muhimu: unachagua level kulingana na mahitaji ya mfanano ya data (data muhimu/ya kifedha inaweza kuhitaji Serializable; programu nyingi hutumia Read Committed kama uwiano).

Kujua kwamba utekelezaji hutofautiana (MVCC dhidi ya locking) pia kunahusika.

Kwa kuwa concurrency ni ya asili katika programu halisi za hifadhidata, na kwa kuwa isolation levels husimamia uwiano muhimu kati ya mfanano na utendaji (na chaguo mbaya zikisababisha aidha kasoro za data au matatizo ya utendaji), kuelewa isolation levels — kasoro, levels, na uwiano wa mfanano-dhidi-ya-concurrency — ni maarifa ya thamani ya kiwango cha senior kwa kujenga mifumo ya concurrent iliyo sahihi, yenye utendaji na mada ya kawaida ya mahojiano ya hali ya juu inayoonyesha uelewa wa concurrency ya hifadhidata.