Scaling Reliability: SRE Practices for 24/7 Financial Markets

For decades, stock exchanges ran on predictable schedules. Markets opened in the morning, closed in the afternoon, and stayed quiet overnight. That rhythm gave infrastructure teams something rare in high‑traffic systems: built‑in downtime. Critical updates, database migrations, and failover tests could be scheduled after the closing bell without disrupting trades.

That world is starting to fade. The success of cryptocurrency exchanges — which operate globally, 24/7, without maintenance windows — is reshaping expectations. Some traditional stock exchanges are already extending their trading hours. Others are exploring continuous operation. For site reliability engineers (SREs) and DevOps teams, this is more than a scheduling shift — it’s a complete change in how reliability is engineered.

Trading Hours Are Expanding — Fast

This transformation is already under way. Traditional U.S. exchanges like NYSE and Nasdaq still operate from 9:30 a.m. to 4:00 p.m. ET, with pre‑market sessions starting around 4:00 a.m. and after‑hours trading running until about 8:00 p.m. ET.

But the NYSE is planning to extend trading on its electronic Arca venue to 22 hours per weekday, stretching from 1:30 a.m. to 11:30 p.m. ET, pending regulatory approval. Similarly, Cboe is pursuing 24-hour, five-day trading via its EDGX Equities Exchange to meet growing demand from Asia-Pacific markets.

Why is this happening? Exchanges are eager to attract global investors — especially from Asia and the Middle East — who operate outside U.S. hours. Extended access boosts liquidity, trading volumes, and positions exchanges as more modern. Investors benefit too, gaining the ability to react instantly to news or earnings without waiting for the next day’s open.

This trend is global. The London Stock Exchange is exploring longer hours to better serve international and retail traders, inspired by crypto’s pace. In Asia, while many exchanges still use traditional hours, international demand is increasing pressure to expand.

The direction is clear: downtime is shrinking, and expectations for near‑constant access are becoming the norm — meaning maintenance windows, once taken for granted, are vanishing.

Why It Matters for Reliability

Downtime in financial markets isn’t theoretical — even major exchanges have suffered consumer‑visible outages. On June 3, 2024, a technical glitch at the NYSE triggered trading halts for about 40 symbols, including Berkshire Hathaway Class A shares, which briefly plunged nearly 99% before trades were voided. The issue stemmed from a faulty software update on the data processing side and was resolved by switching to a backup center. This incident underscored how even short glitches can shake market confidence.

As exchanges push toward extended hours, the operational stakes grow. Cryptocurrency platforms regularly add and remove tokens, demanding rapid adaptability. While stock exchanges don't change listings as frequently, they still face bursts of activity — such as IPOs, mergers, or macroeconomic surprises — that can overwhelm systems without warning.

These shocks can cause sudden traffic spikes, especially around earnings releases or geopolitical news. Without advanced capacity planning, predictive auto-scaling, and protective strategies like rate limiting, systems may slow, orders might be delayed, or outages could occur.

In finance, SRE teams also must ensure zero data loss, maintain transactional integrity, and implement disaster recovery systems to handle failures gracefully. Post-incident, they need transparent, timely communication with traders, regulators, and investors to regain trust.

Reliability engineering in this environment is more than uptime monitoring — it's about building systems resilient to constant change, unpredictable demand, and evolving risk.