# Model tests optical-network scheduling for AI data centers

> Researchers affiliated with Hong Kong Polytechnic University and other institutions in China modeled optical circuit switching alongside network scheduling to shift AI data-center training toward renewable-energy windows. In a controlled five-center scenario, modeled grid-purchase cost was 12.50% lower and local renewable-energy use rose by 2.724 percentage points.

- Canonical URL: https://newsverum.com/en/news/optical-network-ai-data-centers-2026
- Publisher: Newsverum
- Section: Technology
- Published: 2026-10-13T01:40:12+00:00
- Updated: 2026-10-13T01:40:12+00:00
- Source review period: 2026-10-09
- Other editions: [Türkçe](https://newsverum.com/tr/news/optical-network-ai-data-centers-2026.md), [العربية](https://newsverum.com/ar/news/optical-network-ai-data-centers-2026.md), [Français](https://newsverum.com/fr/news/optical-network-ai-data-centers-2026.md), [Español](https://newsverum.com/es/news/optical-network-ai-data-centers-2026.md), [Deutsch](https://newsverum.com/de/news/optical-network-ai-data-centers-2026.md), [Português](https://newsverum.com/pt/news/optical-network-ai-data-centers-2026.md)

*The source does not state the event date; the period shown is the source review period.*

The scheduling method jointly considers where training runs, how long checkpoint data take to cross the network and when optical circuits are reconfigured. The researchers used California ISO electricity series, but transformed and redistributed renewable profiles across sites; checkpoint sizes and data-center capacities were also assumptions.

In this modeled case of five AI data centers, 16 training records and matched optical resources, dynamic optical switching reduced real-time grid-purchase cost by 12.50% and increased local renewable-energy utilization by 2.724 percentage points versus static operation. These are not savings measured at operating data centers; the authors say benefits depend on where energy opportunities occur and on network-transfer bottlenecks.

## Data

### Results from the five-center modeled case

These are results from a controlled model comparison, not savings measured at operating data centers.

Methodology: The study modeled five AI data centers, 16 training records and assumed checkpoint-transfer sizes using California ISO electricity data and transformed renewable-energy/price profiles. Dynamic optical switching was compared with static circuit operation using the same optical resources.

| Comparison outcome | Real-time grid-purchase cost reduction | Increase in local renewable-energy utilization |
|---|---|---|
| Dynamic optical switching versus static optical operation in the controlled five-center case | 12.5 | 2.724 |

Sources: <https://arxiv.org/abs/2610.13061> · Download data: [JSON](https://newsverum.com/api/v1/articles/optical-network-ai-data-centers-2026/data/five-site-constructed-case?lang=en&format=json) · [CSV](https://newsverum.com/api/v1/articles/optical-network-ai-data-centers-2026/data/five-site-constructed-case?lang=en&format=csv)

## Sources

- [arXiv](<https://arxiv.org/abs/2610.13061>) (Source checked: 2026-10-13; Published: 2026-10-09)

---

Cite the canonical URL above. [Editorial standards and corrections](https://newsverum.com/en/editorial)