Reducing RAM usage

Hi Gianluca,

Reducing memory usage in your algorithm can be achieved in several ways. Here are some suggestions based on the information extracts:

1. Minimize your universe and historical requests: As suggested by Nico Xenox in the forum discussion, reducing the frequency of rebalancing or the number of historical requests can help manage memory usage.

2. Optimize your algorithm: Before considering a higher live trading host with more memory capacity, it's essential to optimize your algorithm first to ensure efficient usage of resources.

3. Filter your universe more aggressively: As mentioned by Mia Alissi in the same forum discussion, you can try filtering your universe more aggressively in the CoarseSelectionFunction. For example, you can increase the minimum price or dollar volume thresholds.

4. Manage your data subscriptions: As discussed in the forum discussion by Garrison Whipple, managing your data subscriptions effectively can also help reduce memory usage. Make sure to remove data subscriptions that are no longer needed.

Here's an example of how you can manage your data subscriptions:

def OnSecuritiesChanged(self, changes):
    for security in changes.AddedSecurities:
        self.benzingas[security.Symbol] = self.AddData(BenzingaNews, security.Symbol).Symbol

    for security in changes.RemovedSecurities:
        if security.Symbol in self.benzingas:
            self.RemoveSecurity(self.benzingas[security.Symbol])
            del self.benzingas[security.Symbol]

Remember, each asset in the universe consumes approximately 5MB of RAM, so reducing the number of assets can help with memory usage.

If you still face memory issues after optimizing your algorithm, you may need to consider using a higher live trading host with more memory capacity. However, it's essential to optimize your algorithm first to ensure efficient usage of resources.

For example, i may force it to get less strikes in the option chain, but when i try to do so it results in an error due to missing data. Is that because i subscribe to an option contract but then when spot price moves the algo “unsubscribes” from the same option contract?

yes, it looks like this might happen, in this line:

                current_put_price = data[current_symbol].Close

in OnData handler you are trying to get price data for a symbol from the data slice, but it may not be there, not only because option contract is no longer matching your filter, but simply because there were no trades for this contract in this time slice. So you should always check if data slice actually contains data for your symbol, like this:

if not data.ContainsKey(current_symbol):
    continue