Challenges in Multi-Tenant Web Cache Management

• Anna Lira UFCG
• João Ramalho UFCG
• Ruan Alves UFCG
• Thiago Emmanuel Pereira UFCG
• Francisco Vilar Brasileiro UFCG
• Mariana Mendes VTEX

DOI: https://doi.org/10.5753/sbrc_estendido.2024.1861

Abstract

Although caching is far from being a newly technique for improving system performance, there are still challenges in its operation. For example, multi-tenant web platforms have little support, both in methodology and tooling, to manage their multiple workloads. In particular, it is not clear the trade-offs on choosing between multiple exclusive caches and a single shared cache. This work is the result of a cooperation with an industrial partner that operates a large e-commerce platform. The purpose of the cooperation is to better understand the challenges in multi-tenant cache management. To do this, we characterize this partner’s workload and discuss some strategies that can support the management of multi-tenant caches.

References

Gu, R., Li, S., Dai, H., Wang, H., Luo, Y., Fan, B., Basat, R. B., Wang, K., Song, Z., Chen, S., Wang, B., Huang, Y., and Chen, G. (2023). Adaptive online cache capacity optimization via lightweight working set size estimation at scale. In USENIX ATC 2023, pages 467–484.

Huang, S., Wei, Q., Feng, D., Chen, J., and Chen, C. (2016). Improving flash-based disk cache with lazy adaptive replacement. ACM Trans. Storage, 12(2):8:1–8:24.

Lira, A., Alves, R., Pereira, T. E., Morais, F., Ramalho, J., and Mendes, M. (2024). No clash on cache: Observations from a multi-tenant ecommerce platform. In Proceedings of the 2024 ACM/SPEC International Conference on Performance Engineering.

Xiang, X., Ding, C., Luo, H., and Bao, B. (2013). HOTL: a higher order theory of locality. In ASPLOS 2013, pages 343–356. ACM.

Yang, T., Pollen, S., Uysal, M., Merchant, A., Wolfmeister, H., and Khalid, J. (2023). Cachesack: Theory and experience of google’s admission optimization for datacenter flash caches. ACM Trans. Storage, 19(2):13:1–13:24.