[speedplane]

From the article: "Here, we simulated a multi-tenant e-commerce application where each tenant represented a product category and maintained its own index. For this benchmark, we built 50K indices (or products), which each stored up to 500 documents (or items), for a total of 25 million indices. RediSearch built the indices in just 201 seconds, while running an average of 125K indices/sec. However, Elasticsearch crashed after 921 indices and clearly was not designed to cope with this load."

No sane elasticsearch engineer would make a new index for each product. They would just have a single index with a product_id field for each sub-item. If you needed product level information, you would create a second index for that. You'd use two indexes not O(#Product) indexes.

They just created a botched benchmark by using ES incorrectly. It's like driving a car backwards and then complaining it has poor max speed. ES could easily handle this type of problem if done correctly.