In the first case it adds the label to the node record and the node to the label-node-index which it then uses for search
In the second case it searches all nodes with that label, has to load their properties and check the property for the value on each node So in my experience the first approach is faster But still a global operation Sent from mobile device Am 02.05.2014 um 15:27 schrieb Michael Azerhad <[email protected]>: > Suppose a "Notification" entity. > To keep thing simple, let's assume a single property: "readFlag", that could > be true if the notification was already read by the client, or false if it is > new or not checked by the client yet. > > What is the difference in term of performance between those both ways to > retrieve read notifications?: > > Set a label on the "Notification" node called "ReadNotification" and query > them using: MATCH(r:ReadNotification) return r > Index the "readFlag" property of the "Notification" node without needing a > label. Query would be: MATCH(n:Notification) WHERE n.read = true return n > > > Are both ways very similar? Which one should I choose? > > Thanks a lot, > > Michael > -- > You received this message because you are subscribed to the Google Groups > "Neo4j" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
