A real-time weather application ingests 50,000 sensors sending 10 readings/second (timestamp + value). Which storage is appropriate for low-latency, high-throughput time-series ingestion?

Google Cloud Bigtable is ideal for this scenario due to its high throughput and low-latency capabilities for large analytical and operational workloads, especially time-series data. It is a NoSQL wide-column store designed for petabyte-scale data with consistent sub-10ms latency. The 50,000 sensors sending 10 readings per second generate a massive stream of time-series data (500,000 data points/second) that Bigtable can efficiently ingest and query. Google BigQuery is an excellent data warehouse for analytics but is not optimized for high-volume, low-latency individual record ingestion. Google Cloud SQL is a relational database better suited for structured transactional data, not the extreme scale and velocity of this time-series workload. Google Cloud Storage is an object storage service, good for archiving large files, but not designed for real-time, low-latency data ingestion and querying of individual sensor readings.