GraphScope
GraphScope copied to clipboard
alibaba
🔨 🍇 💻 🚀 GraphScope: A One-Stop Large-Scale Graph Computing System from Alibaba | 一站式图计算系统
**Is your feature request related to a problem? Please describe.** 1. Create a cluster with N>1 nodes 2. Start a session 3. Start any process such as add edges. 4....
**Describe the bug** When loading datagen-8_5-fb and datagen-9_0-fb, if we add the vertices first and then add edges, the session crashes. **To Reproduce** .e file is converted to csv. Output...
- Now, the edge id may be equal to the vertex in the result of gremlin query, which may cause some problems for graph visualization. - It would be better...
Currently, out `make graphscope` will trigger the `make install`, which is not conform to the common mindset. To be clear, the `make` should not involved any `install` phase, it can...
Reported by our users, e.g., `vineyard_shared_mem` is not clear, users may be confused about whether it means the shared_mem on a node or the total amount on the cluster. Suggestion:...
As we discussed before, HDFS will only be used in `data_loader` module for store. Hence we may need to remove HDFS in many places. - [ ] dependencies in poms...
as titled. Proposal: maybe a token/secret configured/generated on the k8s service, and then the exposing service via grpc should check the authentication? refs: https://kubernetes.io/docs/reference/access-authn-authz/authentication/ https://grpc.io/docs/guides/auth/
as titled, proposed by Longbin. Needs to further discuss its pros and cons.
GraphScope builtin `Kcore` algorithm can only run on undirected graph currently. Add support of directed graph in `Kcore`.
