Zanjirlararo tahlil

(Bo’lim: Ekotizim va Tarmoq)

1) Zanjirlararo tahlil nima va nima uchun zarur

• Qimmatlilik va faollikning yagona hisobi: hajmlar, likvidlik, komissiyalar, retenshn.
• Ko’priklar va P2P aloqalarini kuzatish: yakunlash, laglar, reorg/challenge hodisalari.
• Trafik va konversiya atributiyasi: cheyn → cheyn, kanal → mahsulot.
• Xavf va komplayens: AML, sanksiyalar, xulq-atvor frodlari, mohiyatni identifikatsiya qilish.
• Qarorlar qabul qilish: OKR/budjetlar, limitlar, yangilanishlar va likvidlik reglamentlari.

2) Ma’lumotlar manbalari va hodisalar (kanonik ro’yxat)

1. Zanjir/reyestrlar: bloklar, tranzaksiyalar, voqealar loglari, aqlli-kontraktlarning holatlari.
2. Ko’priklar: buyurtmalar, kvitansiyalar, dalillar (light/optimistic/ZK), yakunlash maqomi.
3. To’lov provayderlari/QQS: tekshiruvlardan o’tish, limitlar, to’lovlar maqomi.
4. Mahsulot hodisalari: onbording, depozitlar/stavkalar/xulosalar, o’yin va xulq-atvor metrikalari.
5. P2P-transport: Pub/Sub kvitansiyalari, RPC-muvaffaqiyat, latency.
6. Ma’lumotnomalar: tarmoqlar, aktivlar, decimals, chainId, shartnoma manzillari, SDK versiyalari.

3) Ma’lumotlar arxitekturasi (oqimlar va omborlar)

Ingest (striming): nod/indeks konnektorlari, webhooks ko’priklar, operatsion ma’lumotlardan CDC.
Xom qatlamlar (Bronze/Raw):’observed _ at’belgisi va manba meta-ma’lumotlari bo’lgan o’zgarmas partiyalar.
Tozalash/normallashtirish (Silver): dedup, semantik boyitish, taymzonlarni tekislash, aktivlar mappingi.
Yadro modellari (Gold/Core): unifikatsiyalangan faktlar’transfers’,’bridges’,’onchain _ events’,’kyc _ status’,’payouts’.
Vitrinalar (Marts): moliya (GTV/TVL/Take Rate), mahsulot (retenshn/huni), xavf (skoring), operatsiya (SLO).
Kesh/Serve: Dashbordlar va API uchun OLAP/HTAP va/tx manzillari boʻyicha alohida qidiruv.

Transport: Kafka/Pulsar (idempotentlik ustidagi exactly-once semantics), xomashyo uchun obyekt ombori, tahlillar uchun parket/kolonnali formatlar.

4) Yakunlash, reorglar va idempotentlik

Holatlar:’observed’→’confirmed (k)’→’finalized’→’invalidated (reorg)’.
Tasdiqlash qoidasi (K-confirmations): tarmoq/aktiv turi boʻyicha sozlanadi.
Optimistic/Challenge: ko’priklar uchun «bahsli» maqomni qo’llab-quvvatlash.
Idempotentlik:’idempotency _ key = chainId’block’tx’logIndex’topic’(yoki foydali yuk xeshi).
Qayta oʻynash (replay): rejali backfill va indekser almashtirilganda tiklash.

5) O’ziga xoslik va mohiyat modeli (entity resolution)

Manzil → Aktor: manzil, kalitlar, hamyonlar, hisob raqami/tashkilot/provayder.
Kross-zanjirli grafa: manzillarning bir egasi tomonidan bog’lanishi (evristik, imzo, onbording-ma’lumotlar).
Ishonch darajasi: hard-link (KYC, on-chain imzo), soft-link (xulq-atvor korrelyatsiyalari).
Taxalluslashtirish: tahlilda PII oʻrniga barqaror identifikatorlarni (PID) saqlang.

6) Voqealarning birxillashtirilgan sxemasi (soddalashtirilgan)

yaml event:
id: string # global UUID observed_at: timestamp # when they saw chain_id: string # 'eth-mainnet', 'solana-mainnet',...
block_height: long tx_hash: string log_index: int event_type: string    # transfer    bridge. lock    bridge. mint    kyc. pass    payout. done...
status: string      # observed    confirmed    finalized    invalid actor_src: string # address/peer-id/source organization actor_dst: string # address/peer-id/destination organization asset: string # canonical symbol (e. g., USDC), + decimals amount: decimal usd_value: decimal # rate normalization at the observed_at bridge_ref: string # link with the application/receipt of the metadata bridge: object # network/contract/version/gac/fee, etc.
idempotency_key: string

7) Aktivlar va narxlarni normallashtirish

Aktivlarning kanonik ma’lumotnomasi: ramz, decimals, chain mapping, kontrakt manzillari.
FX normallashtirish: tarixiy kurslar va «observed _ at» taymstempi bo’yicha aktivlar narxlari.
Multi-aktiv bandlalar: «o’ralgan» va nativ aktivlarni guruhlang.

8) Kalit metriklar va vitrinalar

8. 1 Moliya va likvidlilik

Tarmoqlar/aktivlar/ko’priklar bo’yicha GTV (Gross Transaction Volume).
TVL va Net Flow ko’priklar va hovuzlar bo’ylab.
Take Rate/hajm uchun komissiya; Transfer uchun Cost-to-Serve.
Payout SLA Hit Rate, Finality p50/p95, Pending Backlog.

8. 2 Mahsulot va foydalanuvchi

Cross-chain MAU/DAU (dedup по PID),

Retention D1/D7/D30 multi-cheyn faolligini hisobga olgan holda,

Funnel: kirish tarmog’i → ko’prik → maqsadli mahsulot → harakat.
QoT (trafik sifati): anti-froddan keyingi trafik validi.

8. 3 Tavakkalchilik va komplayens

Fraud/Dispute Rate, High-Risk Score%, Sanctions Hit%.
Tarjimalar patternlari bo’yicha anomaly rate, velocity-chek, clustering.
KYB/KYC Pass% va tayminglar.

8. 4 Operatsion va SLO

Bridge Success-Rate, p95 Finality, Relay Availability,

Reorg/Challenge events, Error budget burn.

9) SQL/psevdo-so’rovlar namunalari

Bir juft zanjir bo’yicha GTV

sql
SELECT src. chain_id AS src_chain,
dst. chain_id AS dst_chain,
date_trunc('day', e. observed_at) AS d,
SUM(e. usd_value) AS gtv_usd
FROM events e
JOIN bridges b ON e. bridge_ref = b. id
JOIN networks src ON b. src_chain_id = src. id
JOIN networks dst ON b. dst_chain_id = dst. id
WHERE e. status = 'finalized' AND e. event_type IN ('bridge. lock','bridge. mint','transfer')
GROUP BY 1,2,3;

Cross-chain retention D7

sql
WITH first_touch AS (
SELECT pid, MIN(observed_at) AS t0
FROM product_events
WHERE event IN ('signup','first_deposit')
GROUP BY pid
),
week_activity AS (
SELECT DISTINCT pid
FROM product_events pe
JOIN first_touch ft USING(pid)
WHERE pe. observed_at BETWEEN ft.t0 + INTERVAL '1 day'
AND ft.t0 + INTERVAL '7 day'
)
SELECT 100. 0 COUNT() / (SELECT COUNT() FROM first_touch) AS d7_retention_pct
FROM week_activity;

SLO ko’prigi uchun vitrin

sql
SELECT date_trunc('hour', observed_at) AS h,
100. 0 SUM(CASE WHEN status='finalized' THEN 1 END)/COUNT() AS success_rate,
percentile_cont(0. 95) WITHIN GROUP (ORDER BY (finalized_at - observed_at)) AS p95_finality_min,
SUM(CASE WHEN challenge_event THEN 1 END) AS challenges
FROM bridge_events
WHERE observed_at >= now() - INTERVAL '7 days'
GROUP BY 1;

10) Atributsiya va ko’p kanalli yo’l

Tarmoq, ko’prik va mahsulot manbai uchun tarozilar bilan last-touch/position-based modeli.
UTM → On-chain: kliklarni/referallarni onbordingning oncheyn manzili bilan bogʻlang (rozilik bilan).
Assotsiativ modellar: Shapley/Markov murakkab yo’llar uchun «tarmoq → ko’prik → mahsulot».

11) Anti-frod va xulq-atvor signallari

Grafik belgilar: umumiy kontragentlar, doiraviy tarjimalar, tez aylanish.
Velocity-limitlar va anomaliyalar: portlashlar, summalarni «maydalash», tungi klasterlar.
Ko’priklardagi firibgarlik sxemalari: qayta berish, KYCni chetlab o’tishga urinish, likvidli sendvich-patternlar.
Modellar: gradient busting/graph-embeddings; hodisalarni belgilashda oʻrgating.

12) Maxfiylik va komplayens (privacy-by-design)

PII minimallashtirish: toʻgʻridan-toʻgʻri identifikatorlar oʻrniga PID, tokenizatsiya.
Data residency: mintaqalar bo’yicha partiyalashtirish, «tinch/yo’lda» shifrlash.
Olib tashlash huquqi: isbotlanishi mumkin boʻlgan tombstone/redaction-hodisalar.
Kirish va audit: rolli ACL, oʻqish jurnallari, tekshirish uchun imzolangan hisobotlar.

13) Analitik payplaynlar uchun SLI/SLO

• Freshness (mediana laga’observed _ at’dan Goldda paydo bo’lgunga qadar),
• Completeness (K-confirmations kutgan holda teshiksiz voqealar%),
• Correctness (sxemalar/qoidalar validatsiyasidan o’tgan voqealar%),
• Reorg handling success (% to’g "ri nogironlik/takrorlash),
• Serve latency (vitrinalar/dashbordlarga p95 so’rovlar).

• Freshness p95 ≤ 3 min (striming), ≤ 15 min (batch).
• Completeness ≥ 99. 7%, Correctness ≥ 99. 9%.
• Reorg handling success ≥ 99. 9%.
• Serve p95 ≤ 500 ms (asosiy vitrinalar).

14) Ma’lumotlar kuzatilishi va lineage

Data Lineage: dashborddan xom hodisagacha (column-level).
Sifat signallari: completeness, uniqueness, referential integrity, schema drift.
Alertlar: «jim nosozliklar» (yangi ma’lumotlar yo’q), taqsimot sakrashlari,’unknown’maydonlarining o’sishi.

15) Dashbordlar (shablonlar)

• Success-Rate, p95 Finality, Relay Availability, Challenge/Reorg, backlog, error budget burn.

• TVL, Net Flow per chain, cost-per-transfer, utilization, sug’urta fondi.

• MAU/DAU (dedup), cross-chain retention, kanal voronkalari, QoT.

• Fraud/Dispute Rate, sanctions hits, high-risk share, protsesslar tezligi.

16) Operatsion reglamentlar va playbook

Hodisa: lag tazelik> SLO

Konnektorlar/indekserlarni tekshirish, zaxiraga o’tish, degradatsiya rejimini yoqish (vitrinalar «oxirgi tugallangan» ni ko’rsatadi), manbaning egasiga eskalat qilish.

Hodisa: reorg/challenge

K-confirmations/nizo oynasini ko’paytirish, katta mablag’lar uchun «delayed finalization» ni kiritish, ko’prik/operatorlarni xabardor qilish.

Hodisa: valyuta/aktivlar tafovuti

Ta’sir ko’rsatilgan juftliklarni muzlatish, ma’lumotnomani qaytarish, USD-normallashtirishni qayta hisoblash, hisobotni e’lon qilish.

Hodisa: Fraud/Dispute sakrash

Limitlarni/skoringni kuchaytirish, high-riskni qo’lda yig’ish, modelni yangi patternda o’rganish.

17) Konfiguratsiya namunasi (psevdo-YAML)

Tarmoqlar boʻyicha yakunlash oynalari

yaml finality:
eth-mainnet: 12  # блоков polygon: 256 solana: "optimistic: 32 slots"
optimistic-bridge: { challenge_minutes: 20 }
zk-bridge: { proof_time_sla: 180 }

Idempotentlik va dedup qoidalari

yaml dedup:
key_template: "${chain_id}    ${block_height}    ${tx_hash}    ${log_index}    ${event_type}"
ttl_hours: 48

Payplaynlarning SLO

yaml pipelines:
ingest_stream:
freshness_p95_min: 3 completeness_min_pct: 99. 7 gold_build:
correctness_min_pct: 99. 9 reorg_success_min_pct: 99. 9

18) Joriy etish chek-varaqasi

1. Manbalar, sxemalar, yakunlash oynalari va egalarini belgilang.
2. Idempotentlik va reorg-handling (states + replay) ni yoqing.
3. Modellar yadrosini quring (transfers/bridges/onchain_events/kyc/payouts).
4. Aktivlar va FX ma’lumotnomalarini moslashtiring.
5. SLI/SLO payplaynlar va dashbordlarni aniqlang.
6. Entity resolution va privacy-by-design dasturlarini amalga oshiring.
7. Anti-frod skoringlar va hodisalar reglamentini yoqing.
8. Tarixiy reorg/challenge keyslarida backfill va testlarni o’tkazing.
9. Sxemalar, metrik vazn va manbalarni muntazam tekshirib turing.

19) Glossariy

Finality - holat/hodisaning qaytarilmasligi.
Reorg - bloklarning bir qismini bekor qilishga olib keladigan zanjirni qayta yig’ish.
Challenge period - optimistik modellarda bahslashish oynasi.
Entity resolution - yagona mohiyatdagi manzillar/akkauntlarni taqqoslash.
GTV/TVL - tranzaksiyalar hajmi/bloklangan qiymat.
Completeness/Freshness/Correctness - ma’lumotlar sifatining bazaviy metrikasi.

Xulosa: zanjirlararo tahlillar - bu shunchaki metrika ma’lumotlari emas, balki boshqariladigan intizom: voqealarning yagona sxemasi, to’g’ri yakunlanishi, barqaror payplaynlar, maxfiylik, anti-frod va tushunarli vitrinalar. Ushbu freymvorga amal qilib, ekotizim haqiqatan ham xom blokdan tortib biznes yechimigacha bo’lgan qiymat, xavf va o’sish nuqtai nazariga ega bo’ladi.