# Update 24 September 2021 - 30 September 2021

This week I am focused on learning about [LES(Light Ethereum subprotocol)](https://hackmd.io/@Nhlanhla/Bk0dDeJ4Y), from what i have gathered so far with limited resources I have come across internet,is that they only download block headers as they appear and fetch other parts of the blockchain on-demand. They provide full functionality in terms of safely accessing the blockchain, but do not mine and therefore do not take part in the consensus process. Full and archive nodes can also support the 'les' protocol besides 'eth' in order to be able to serve light nodes.

[LES](https://hackmd.io/@Nhlanhla/Bk0dDeJ4Y) use Cononial Hash trie structure (specifically Patricia verkle tree) for quick initial syncing and secure on-demand retrieval of the Cononial Hash mapping, block headers and total difficulty values.

The BloomBits data structure optimizes log searching by doing a bitwise transformation that makes it cheaper to retrieve bloom filter data relevant to a specific filter. When searching in a long section of the block history, we are checking three specific bits of each bloom filter per queried address/topic. In order to do that, LES must retrieve a ~550 byte block header per filtered block. BloomBits data is stored in compressed form. The compression algorithm is optimized for sparse input data which contains a lot of zero bytes. Decompression requires knowledge of the decompressed data length.

Any node which takes on a server role in the the LES protocol needs to be able to somehow limit the amount of work it does for each client peer during a given time period. They can always just serve requests slowly if they are overloaded, but it is beneficial to give some sort of flow control feedback to the clients. This way, clients could (and would have incentive to) behave nicely and not send requests too quickly in the first place (and then possibly timeout and resend while the server is still working on them). They could also distribute requests better between multiple servers they are connected to. And if clients can do this, servers can expect them to do this and throttle or drop them if they break the flow control rules.