Vemcache is an in-memory vector database.
Vemcache supports a variety of commands to interact with vector data. Below is a list of available commands and their descriptions:
ping: Check the connection to the Vemcache server. The server responds with “pong” when it receives this command.
insert [values]: Insert a vector into the database. The server generates a unique identifier (UUID) for the vector. Provide space-separated floating-point values as the vector components.
named_insert [key] [values]: Insert a vector into the database with a specified key. Provide the key as a string and space-separated floating-point values as the vector components.
get [key]: Retrieve a vector from the database using its key.
remove [key]: Remove a vector from the database using its key.
knn [key] [k]: Find the k nearest neighbors of a vector. Provide the key of the query vector and the value of k (number of neighbors).
vadd [key1] [key2]: Perform element-wise addition of two vectors. Provide the keys of the two vectors to be added.
vsub [key1] [key2]: Perform element-wise subtraction of two vectors. Provide the keys of the two vectors to be subtracted.
vscale [key] [scalar]: Scale a vector by a scalar value. Provide the key of the vector to be scaled and the scalar value.
vcosine [key1] [key2]: Calculate the cosine similarity between two vectors. Provide the keys of the two vectors to be compared.
dump [filename]: Dump Vemcache DB to a JSON file.
To use Vemcache, connect to the server using a TCP client like telnet or nc. Once connected, you can send commands to interact with the server.
Use telnet to connect to Vemcache
telnet 0.0.0.0 7070
Or use nc
nc 0.0.0.0 7070
Here are some example commands to interact with Vemcache:
# Insert a vector with values 0.0, 1.0, 2.0
insert 0.0 1.0 2.0
# Insert a vector with key "my_vector" and values 0.0, 1.0, 2.0
named_insert my_vector 0.0 1.0 2.0
# Get the vector associated with key "my_vector"
get my_vector
# Remove the vector associated with key "my_vector"
remove my_vector
# Find the 3 nearest neighbors of the vector with key "query_vector"
knn query_vector 3
# Add vectors with keys "vector1" and "vector2"
vadd vector1 vector2
# Subtract vectors with keys "vector1" and "vector2"
vsub vector1 vector2
# Scale vector with key "vector1" by scalar 2.0
vscale vector1 2.0
# Calculate cosine similarity between vectors with keys "vector1" and "vector2"
vcosine vector1 vector2
# Dump vemcache db
dump vemcache.json