The 5th KIR: S2WLAB_”S2_EYEZ”_Progress Report(3)_Final

Completed
3

Thank you for your interest.

  1. Please check this out (The 5th KIR: S2WLAB_”S2_EYEZ”_Progress Report(2)).

  2. A graph DB has advantages to explorer transaction data, but general purpose graph DBs force strong ACID properties. Thus, we newly designed and implemented a graph database, not commercial or open source ones. Our first deliverable details the internal graph DB (The 5th KIR : S2WLAB_S2_EYEZ_Progress Report(1))

  3. As you can see the evaluation section, it depends on how many nodes are visited. S2_EYEZ is more faster than other storage-oriented block explorers.

  4. After importing transaction data from the blockchain data, S2_EYEZ manages the data in a relational DB. It happens once.

  5. Yes. By default, BFS is preferred.

  6. Yes. Please see the details in our first deliverable.

  7. a) Sequential search means that reading transaction data including address/transaction nodes with their edges. b,c) I think the question B and C are similar. There are two main things to affect the performance: i) How many transactions (from the newly mined block or queries) rise cache-miss and ii) How much memory S2_EYEZ can use. All depend on data in blocks or queries. If users want to guarantee the constant time of them, one possible strategy is S2_EYEZ forcely keeps specific addresses/transactions to memory.

Please leave the comments, if you have further questions :slight_smile: